CN116183623B - Intelligent wafer surface defect detection method and device - Google Patents
Intelligent wafer surface defect detection method and device Download PDFInfo
- Publication number
- CN116183623B CN116183623B CN202310008991.7A CN202310008991A CN116183623B CN 116183623 B CN116183623 B CN 116183623B CN 202310008991 A CN202310008991 A CN 202310008991A CN 116183623 B CN116183623 B CN 116183623B
- Authority
- CN
- China
- Prior art keywords
- wafer
- detection
- photographing module
- mode
- light source
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 176
- 230000007547 defect Effects 0.000 title claims abstract description 83
- 238000005498 polishing Methods 0.000 claims abstract description 37
- 238000012545 processing Methods 0.000 claims abstract description 23
- 239000000969 carrier Substances 0.000 claims abstract description 10
- 238000004422 calculation algorithm Methods 0.000 claims description 10
- 230000033001 locomotion Effects 0.000 claims description 5
- 238000011109 contamination Methods 0.000 claims description 4
- 210000000078 claw Anatomy 0.000 claims description 3
- 230000011218 segmentation Effects 0.000 claims description 3
- 230000009466 transformation Effects 0.000 claims description 3
- 239000003814 drug Substances 0.000 claims description 2
- 239000007788 liquid Substances 0.000 claims description 2
- 238000000034 method Methods 0.000 abstract description 22
- 235000012431 wafers Nutrition 0.000 description 190
- 230000000007 visual effect Effects 0.000 description 11
- 230000008569 process Effects 0.000 description 10
- 238000010586 diagram Methods 0.000 description 9
- 238000004519 manufacturing process Methods 0.000 description 8
- XUIMIQQOPSSXEZ-UHFFFAOYSA-N Silicon Chemical compound [Si] XUIMIQQOPSSXEZ-UHFFFAOYSA-N 0.000 description 7
- 229910052710 silicon Inorganic materials 0.000 description 7
- 239000010703 silicon Substances 0.000 description 7
- 238000001179 sorption measurement Methods 0.000 description 6
- 230000000694 effects Effects 0.000 description 5
- 239000013078 crystal Substances 0.000 description 4
- 238000007781 pre-processing Methods 0.000 description 4
- 238000004891 communication Methods 0.000 description 3
- 238000004590 computer program Methods 0.000 description 3
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 230000002950 deficient Effects 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 238000007689 inspection Methods 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- XLOMVQKBTHCTTD-UHFFFAOYSA-N Zinc monoxide Chemical compound [Zn]=O XLOMVQKBTHCTTD-UHFFFAOYSA-N 0.000 description 2
- 101150038956 cup-4 gene Proteins 0.000 description 2
- 238000005520 cutting process Methods 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 238000001914 filtration Methods 0.000 description 2
- 238000005286 illumination Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 229920002120 photoresistant polymer Polymers 0.000 description 2
- 239000004065 semiconductor Substances 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 238000003860 storage Methods 0.000 description 2
- JBRZTFJDHDCESZ-UHFFFAOYSA-N AsGa Chemical compound [As]#[Ga] JBRZTFJDHDCESZ-UHFFFAOYSA-N 0.000 description 1
- 208000009043 Chemical Burns Diseases 0.000 description 1
- 208000018380 Chemical injury Diseases 0.000 description 1
- 229910002601 GaN Inorganic materials 0.000 description 1
- 229910001218 Gallium arsenide Inorganic materials 0.000 description 1
- JMASRVWKEDWRBT-UHFFFAOYSA-N Gallium nitride Chemical compound [Ga]#N JMASRVWKEDWRBT-UHFFFAOYSA-N 0.000 description 1
- GPXJNWSHGFTCBW-UHFFFAOYSA-N Indium phosphide Chemical compound [In]#P GPXJNWSHGFTCBW-UHFFFAOYSA-N 0.000 description 1
- 230000002159 abnormal effect Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000002146 bilateral effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000005229 chemical vapour deposition Methods 0.000 description 1
- 239000011248 coating agent Substances 0.000 description 1
- 238000000576 coating method Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- PMHQVHHXPFUNSP-UHFFFAOYSA-M copper(1+);methylsulfanylmethane;bromide Chemical compound Br[Cu].CSC PMHQVHHXPFUNSP-UHFFFAOYSA-M 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 229910003460 diamond Inorganic materials 0.000 description 1
- 239000010432 diamond Substances 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 229910052732 germanium Inorganic materials 0.000 description 1
- GNPVGFCGXDBREM-UHFFFAOYSA-N germanium atom Chemical compound [Ge] GNPVGFCGXDBREM-UHFFFAOYSA-N 0.000 description 1
- 238000010438 heat treatment Methods 0.000 description 1
- 238000003780 insertion Methods 0.000 description 1
- 230000037431 insertion Effects 0.000 description 1
- 238000001259 photo etching Methods 0.000 description 1
- 238000000206 photolithography Methods 0.000 description 1
- 238000007517 polishing process Methods 0.000 description 1
- 229910021420 polycrystalline silicon Inorganic materials 0.000 description 1
- 229920005591 polysilicon Polymers 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- HBMJWWWQQXIZIP-UHFFFAOYSA-N silicon carbide Chemical compound [Si+]#[C-] HBMJWWWQQXIZIP-UHFFFAOYSA-N 0.000 description 1
- 229910010271 silicon carbide Inorganic materials 0.000 description 1
- 239000002210 silicon-based material Substances 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
- 229940095676 wafer product Drugs 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
- 239000011787 zinc oxide Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/95—Investigating the presence of flaws or contamination characterised by the material or shape of the object to be examined
- G01N21/9501—Semiconductor wafers
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8806—Specially adapted optical and illumination features
Landscapes
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
- Testing Or Measuring Of Semiconductors Or The Like (AREA)
Abstract
The invention discloses an intelligent detection method and device for wafer surface defects, which belong to the technical field of wafer defect detection, and the detection method comprises the following steps: the wafer and the wafer vision photographing module are respectively arranged on the corresponding carriers; controlling the wafer or the wafer vision photographing module to move to a first detection initial relative position; the wafer vision photographing module is used for collecting the surface image of the wafer in a detection mode, and the method comprises photographing by path points in a preset detection path in the detection mode; controlling the wafer or the wafer vision photographing module to move to a second detection initial relative position; the wafer vision photographing module is used for collecting the surface image of the wafer in the two detection modes, and photographing is carried out on path points by path points in the preset detection paths in the two detection modes; and processing the shooting images obtained in the first detection mode and the second detection mode to generate detection results. The invention can accurately and efficiently detect various defects on the surface of the wafer by a polishing mode with high degree of freedom.
Description
Technical Field
The invention relates to the technical field of wafer defect detection, in particular to an intelligent detection method and device for wafer surface defects.
Background
The wafer refers to a wafer used for manufacturing a semiconductor circuit, and common materials include silicon, germanium, gallium arsenide, silicon carbide, aluminum nitride, zinc oxide, diamond, indium phosphide, gallium nitride and the like; corresponding characteristics are obtained by doping different elements thereto; taking silicon material as an example, after high-purity polysilicon is dissolved, silicon crystal seed crystal is doped, and the preparation of a silicon rod is completed by a pulling method; the silicon crystal bar is ground, polished and sliced to form a silicon wafer, namely a wafer; in the wafer manufacturing process, a series of working procedures and steps such as chemical vapor deposition, optical development, chemical mechanical polishing, single crystal pulling, slicing, lapping, polishing, layering, photoetching, doping, heat treatment, scribing and the like are performed; because of the great number of semiconductor wafer processing technologies, during the processing technological process and the carrying process, the surface defects of the wafer are easily caused due to abnormal actions of an instrument, improper operation of workers and the like in the processing procedure.
Defects caused in the carrying process are sucker marks which are left on the surface of the wafer due to improper pressure adjustment when the sucker carries the wafer; the device is used for clamping and fixing or processing the wafer, so that the device can produce the claw cracks, broken edges, scratches and scratches on the wafer; contamination, pinholes and chemical burn on the wafer surface due to uncleanness of photoresist removal after photoresist coating; and the uneven edge removal defect is generated due to uneven cutting in the wafer cutting process.
The defects cause more serious defects to the photolithography process after the defective wafer; and the defective wafer enters the next processing and manufacturing procedure, which causes waste of production and manufacturing resources and affects the subsequent wafer processing yield and the overall manufacturing and processing efficiency.
With advances in industrial image sensors, computer vision, and image processing hardware computing, the detection of wafer surface defects by optical inspection equipment and machine vision in combination with process analysis of the cause of the defects has evolved and applied to improve the manufacturing process. At present, the defect detection of the wafer is finished by means of machine vision and computer vision, most cases are fixed polishing, and only single and simple-structure defects on the surface of the wafer can be detected; the real situation of the actual wafer detection industry is that the wafer product is updated and iterated quickly, and the fixed polishing mode cannot meet more and more types of defects for the characteristics of more and more abundant wafer detection processes and material changes and various defect types. The detection mode and the detection method have the advantages of single shooting mode, lower automation degree, difficult collection of defect data sets and poor universality.
Disclosure of Invention
The present invention aims to solve at least one of the technical problems in the related art to some extent. Therefore, the invention provides the intelligent detection method and the intelligent detection device for the wafer surface defects, which finish accurate and efficient detection through a polishing mode with high degree of freedom and a detection algorithm with strong self-adaptability, and are the detection method for the wafer surface defects with high universality, strong reliability and high ductility.
In order to achieve the above object, in a first aspect, the present application provides an intelligent wafer surface defect detection method, including: the wafer and the wafer vision photographing module are respectively arranged on the corresponding carriers;
Controlling the wafer or the wafer vision photographing module to move to a first detection initial relative position;
The wafer vision photographing module is used for collecting the surface image of the wafer in a detection mode, and photographing is carried out on path points by path points under a preset detection path in the detection mode, and circulation is carried out to the end point of the detection path;
Controlling the wafer or the wafer vision photographing module to move to a second detection initial relative position;
the wafer vision photographing module is used for collecting the surface image of the wafer in the two detection modes, including photographing by path points under a preset detection path in the two detection modes, and circulating to the detection path end point;
and performing image detection processing on all the shot images obtained in the first detection mode and the second detection mode to generate a wafer surface defect detection result.
Preferably, the detecting a pattern is performed by polishing the surface of the wafer by the LED diffuse parallel light source under different inclination angles.
Preferably, the detecting two modes are used for polishing the surface of the wafer under different inclination angles through a high-intensity light source with illuminance larger than a first threshold value.
Preferably, polishing the surface of the wafer in the first detection mode and the second detection mode at different inclination angles at least comprises a high polishing angle range and a low polishing angle range, wherein the polishing angle alpha is an included angle between the light source emergent direction and the normal direction of the surface of the wafer, the high polishing angle range is alpha=0 °° -30, and the low polishing angle range is alpha=60 °° -90.
Preferably, the wafer and the wafer vision photographing module are respectively mounted on corresponding carriers, and the control of the wafer or the wafer vision photographing module to move to the first detection starting relative position comprises the steps of adsorbing the wafer to be detected on a sucker on a tail end support of the multi-degree-of-freedom robot, fixedly mounting the wafer vision photographing module on the module support, and controlling the multi-degree-of-freedom robot to drive the wafer to be detected to move to the first detection starting relative position.
Preferably, the wafer and the wafer vision photographing module are respectively mounted on corresponding carriers, and the controlling of the wafer or the wafer vision photographing module to move to the first detection starting relative position includes mounting the wafer vision photographing module on a terminal support of the multi-degree-of-freedom robot, placing the wafer to be tested on the module support, and controlling the multi-degree-of-freedom robot to drive the vision photographing module to move to the first detection starting relative position.
Preferably, the image detection processing is performed on all the shot images obtained in the first mode and the second mode, and the generation of the wafer surface defect detection result includes the steps of preprocessing all the shot images, performing region segmentation processing to cut out the wafer image, and using the optimized YOLOV model to complete the surface defect target detection of the wafer image.
In a second aspect, the present application provides an intelligent wafer surface defect detection apparatus, including:
The wafer visual photographing device comprises a detection workbench, a wafer visual photographing module and a multi-degree-of-freedom robot, wherein the wafer visual photographing module and the multi-degree-of-freedom robot are symmetrically arranged on the detection workbench, the wafer visual photographing module is used for photographing images of defects on the surface of a wafer, and the multi-degree-of-freedom robot is used for driving the wafer or the wafer visual photographing module to perform position movement transformation.
Preferably, the wafer vision photographing module comprises an industrial camera I, an industrial camera II, a high-intensity light source and an LED diffuse parallel light source, wherein the industrial camera I and the LED diffuse parallel light source are used for collecting the wafer surface image in a detection mode, and the industrial camera II and the high-intensity light source are used for collecting the wafer surface image in a detection mode.
Compared with the prior art, the invention has the beneficial effects that:
the intelligent detection method for the surface defects of the wafer can accurately and efficiently detect various defects on the surface of the wafer in a high-degree-of-freedom polishing mode, and can achieve the aim of detecting the surface defects of the wafer in multiple angles, full sizes and full types through the mutual matching of different detection modes.
Additional features and advantages of the application will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the application, the objects and other advantages of the application may be realized and obtained by the construction particularly pointed out in the written description, claims, and drawings.
Drawings
FIG. 1 is a flow chart of an intelligent wafer surface defect detection method according to the present invention;
FIG. 2 is a schematic diagram of a high-angle polishing in an intelligent wafer surface defect detection method according to the present invention;
FIG. 3 is a schematic diagram illustrating a low-angle polishing process in the intelligent detection method of wafer surface defects according to the present invention;
FIG. 4 is a schematic diagram of a first embodiment of an intelligent wafer surface defect inspection apparatus according to the present invention;
FIG. 5 is a schematic diagram of a second embodiment of an intelligent wafer surface defect inspection apparatus according to the present invention;
FIG. 6 is a schematic diagram of a wafer visual photographing module in an intelligent wafer surface defect detecting device according to the present invention;
FIG. 7 is a network configuration diagram of a wafer surface defect intelligent detection method according to the present invention.
In the figure: 1. a module support; 2. a wafer vision photographing module; 3. a wafer; 4. a suction cup; 5. a terminal stent; 6. a multi-degree-of-freedom robot; 7. a detection workbench; 8. a wafer support; 2-1, an industrial camera I; 2-2, an industrial lens I; 2-3, an industrial camera II; 2-4, an industrial lens II; 2-5, high intensity light source; 2-6, LED diffuse parallel light sources; 2-7, a sensor bracket.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
As shown in fig. 1, in a first aspect, a method for intelligently detecting a wafer surface defect according to a first embodiment of the present invention includes:
the wafer 3 and the wafer vision photographing module 2 are respectively arranged on corresponding carriers;
controlling the wafer 3 or the wafer vision photographing module 2 to move to a first detection initial relative position;
the wafer vision photographing module 2 is utilized to collect the wafer surface image in a detection mode, including photographing by path points in a preset detection path in the detection mode, and circulating to the detection path end point;
controlling the wafer 3 or the wafer vision photographing module 2 to move to a second detection initial relative position;
The wafer vision photographing module 2 is utilized to collect the wafer surface image in the two detection modes, including photographing by path points under the preset detection path in the two detection modes, and circulating to the detection path end point;
and performing image detection processing on all the shot images obtained in the first detection mode and the second detection mode to generate a wafer surface defect detection result.
The invention is also provided with a detection algorithm with strong self-adaptability to finish accurate and efficient detection, the method automatically identifies the image type by calculating the average gray value and the gray distribution uniformity, judges the bright and dark fields, uses a histogram equalization and a dynamic bilateral filtering algorithm to carry out denoising sharpening treatment aiming at the bright fields, processes the original image by using a method for removing the over-exploded image aiming at the dark fields, uses a dynamic threshold method to filter the image, and then adds a layer of enhancement feature extractor aiming at the defect contrast difference to identify and extract the defect information on the wafer surface by improving YOLOV, so the self-adaptive algorithm has high universality and strong reliability.
Preferably, the detection mode is a full-size detection mode of diffuse reflection of parallel light on the surface of the wafer, wherein the polishing of the surface of the wafer is performed under different inclination angles through the LED diffuse parallel light sources 2-6. The two detection modes are used for polishing the surface of the wafer under different inclination angles through the high-intensity light source 2-5 with illuminance larger than a first threshold value. The polishing of the surface of the wafer is carried out under different inclined angles in the first detection mode and the second detection mode, wherein the polishing angle alpha is an included angle between the emergent direction of the light source and the normal direction of the surface of the wafer, the high polishing angle range is alpha=0 DEG-30, and the low polishing angle range is alpha=60 DEG-90.
As shown in fig. 2 and fig. 3, the structural schematic diagrams of high-angle polishing and low-angle polishing are respectively shown, the calculation formula is h c＝do ×tan (α), where h c is the calculated characteristic depth of the crack that can be identified, d o is the current crack width, α is the included angle of the light source camera, and the current detected crack and scratch defect of the silicon wafer are counted, and the aspect ratio of the hole is distributed about 2.5, so that the camera is required to face the detected object for low-angle polishing, and the light source and the camera form an included angle α. Experiments prove that the characteristic best performance of the included angle between the emergent direction of the light source and the normal direction of the surface of the wafer is that alpha=60 DEG-90 deg. By applying the law of reflection, the high-angle polishing obtains more uniform illumination effect on the silicon wafer, the included angle between the emergent direction of the light source and the normal direction of the surface of the wafer is in the range of alpha=0 DEG-30 DEG, and the most uniform illumination effect can be obtained through experiments.
Detecting a low-angle lighting mode, wherein the included angle between the emergent direction of the light source and the normal direction of the surface of the wafer is alpha=60 DEG-90 under the low-lighting angle range, and the low-angle lighting mode is detected; under the current form, the collected image can highlight the types of wafer surface defects such as claw cracks, scratches and the like, enhance the contrast between a defect area and a non-defect area, and has better image quality; when the included angle between the outgoing direction of the light source and the normal direction of the wafer surface is alpha=0 DEG-30, the method is to detect a high-angle polishing mode of a mode, and most of the incident light of the light source enters the camera through specular reflection after passing through the surface of the measured object, which is also called a black-and-white pattern polishing mode; under the current form, the acquired image can highlight the contrast of the wafer edge image, and the gray level of the wafer image is more uniform.
In the detection two modes, when the included angle between the emergent direction of the high-intensity light source and the normal direction of the wafer surface is alpha=60 DEG-90, the angle is the low-angle polishing mode of the detection two modes, and under the high-intensity light irradiation, the defect which is not easy to highlight under the diffuse reflection polishing principle condition under the low-intensity light is more sharp; under the current form, the collected image highlights the areas of the surface defects of the wafer such as contamination, pinholes and the like to show bright spots to be converged, and the image quality is better for the adhesion defects such as contamination and pinholes; in the two detection modes, when the included angle between the emergent direction of the high-intensity light source and the normal direction of the wafer surface is alpha=0 DEG-30. The method is a high-angle polishing mode for detecting the two modes, and in the mode, the burned area of the surface of the wafer due to the residual liquid medicine is developed into a bright surface or a shadow under high-intensity light irradiation, so that the area is easier to extract by a subsequent defect detection algorithm.
As shown in fig. 4, in the first embodiment of the present invention, the wafer 3 and the wafer vision photographing module 2 are respectively mounted on corresponding carriers, and controlling the wafer 3 or the wafer vision photographing module 2 to move to the first detection starting relative position includes adsorbing the wafer 3 to be tested on the suction cup 4 on the end support 5 of the multiple degree of freedom robot 6, and fixedly mounting the wafer vision photographing module 2 on the module support 1, and controlling the multiple degree of freedom robot 6 to drive the wafer 3 to be tested to move to the first detection starting relative position. Wherein the multi-degree-of-freedom robot 6 is an intelligent wafer detection cooperative robot with 6 degrees of freedom;
the tail end of the multi-degree-of-freedom robot 6 is provided with a wafer adsorption tail end execution module consisting of a sucker 4 and a tail end bracket 5, and the insertion, the taking and the adsorption of the wafer can be completed through the control of an electromagnetic valve; the wafers are sequentially arranged in the wafer box, a certain gap exists between the wafers, and the wafers are independent of each other and cannot be contacted with each other.
When a wafer is adsorbed, a wafer adsorption end execution module carried by the multi-degree-of-freedom robot 6 moves to a gap between the wafer to be detected and the next wafer, the wafer adsorption end execution module is inserted in the direction parallel to the surface of a sucker 4 of the wafer adsorption end execution module, the posture is adjusted to enable the plane of the sucker 4 to be parallel to the plane of the wafer and to be in a near contact position, the current position is the position of the multi-degree-of-freedom robot 6 for adsorbing the wafer, and the multi-degree-of-freedom robot 6 is in place and sends an in-place signal; the upper computer software system receives a signal sent by the robot, sends an adsorption instruction to the motion controller, and the motion controller controls the electromagnetic valve to respond, and the tail end sucker 4 finishes adsorbing the wafer.
After a wafer to be detected is adsorbed by the tail end sucker 4, the wafer is moved to a first detection area, and after reaching a detection starting position, a robot sends an in-place signal; after the upper computer software system receives the signal, the light source of the wafer detection parallel light detection module is controlled to be turned on, the light source of the wafer detection high-intensity light detection module is turned off, and the detection system enters a detection mode; the tail end sucker 4 always keeps the state that the wafer is absorbed before the detection is finished, the tail end sucker 4 carries the wafer to be detected, and moves according to a preset position path; in the detection process of detecting a mode, the upper computer software system acquires position parameters sent by the robot in real time, and after judging that the robot is in place, the upper computer software system controls the triggering camera to take a picture; shooting to obtain an image, entering an upper computer software system, processing the image through a wafer surface defect detection algorithm, and outputting a result to a wafer surface defect detection result list; then photographing, detecting and processing the path points by path points, and sequentially circulating to the detection path end point;
The end sucker 4 carries the wafer to be tested and moves to a starting position for detecting a mode, and after the end sucker reaches the position, the robot sends an in-place signal; after the upper computer software system receives the robot in-place signal, the light source of the wafer detection parallel light detection module is turned off, the light source of the wafer detection high-intensity light detection module is turned on, and the two detection modes are entered; the robot carrying wafer moves according to a detection path in a second detection area, and at each shooting position point, an upper computer software system acquires position parameters sent by the robot in real time, and after judging that the robot is in place, the upper computer software system controls a trigger camera to shoot; in a wafer surface defect detection algorithm of which the image is transmitted into an upper computer software system, processing the image through the wafer surface defect detection algorithm, and outputting a result to a wafer surface defect detection result list; then, photographing, detecting and processing are carried out on the path points by path points, and the paths are sequentially circulated to the detection path end points; and after the photographing process is finished, traversing the wafer surface defect detection result list by the central control system, judging whether the current wafer surface has defects, and putting the defective wafer and the normal wafer into the corresponding wafer placing bin separately to finish detection.
As shown in fig. 5, the second embodiment of the present invention is different from the first embodiment in that the wafer 3 and the wafer vision photographing module 2 are respectively mounted on the corresponding carriers, and controlling the wafer 3 or the wafer vision photographing module 2 to move to the first detection starting relative position includes mounting the wafer vision photographing module 2 on the end bracket 5 of the multiple degree of freedom robot 6, and placing the wafer 3 to be tested on the module bracket 1, and controlling the multiple degree of freedom robot 6 to drive the vision photographing module to move to the first detection starting relative position. Specifically, in this embodiment, after the end suction cup 4 of the automatic feeding module adsorbs a wafer to be detected, the wafer is placed on a detection station of the wafer support 8, and after a detection station sensor detects that a new wafer to be detected is placed, a signal of completion of feeding is triggered.
Preferably, the performing image detection processing on all the shot images to generate a wafer surface defect detection result includes performing region segmentation processing on all the shot images after preprocessing to cut out the wafer images, and using an optimized YOLOV model to complete surface defect target detection on the wafer images. Wherein preprocessing all captured images includes filtering and histogram equalization.
One reason that the original YOLOV model has poor effect on detecting the surface defects of the wafer is that the contrast between the defects of the wafer and the background is weak, the size of the notch on the surface of the wafer is small, the downsampling multiple of the original YOLOV is large, the deeper feature map is difficult to learn the feature information of the small target, and therefore, the small target detection layer is added to detect the spliced shallow feature map and the deep feature map.
After last upsampling after the original network feature extraction layer, the upsampling and other processes are continuously performed on the feature map, so that the feature map is continuously expanded, and meanwhile, the obtained feature map with the size of 152X152 is subjected to concat fusion with the layer 2 feature map in the backbone network, so that a larger feature map is obtained, and small target detection is performed. For the original YOLOV network structure of each model, such as YOLOV-x with the most complex network structure, the number of layers of CSP use can be repeated for a plurality of times in the network structure, which is similar to deepening of the network structure, but the characteristics of wafer such as unobvious characteristic and background difference, small characteristic size and non-ideal detection effect are not attached. By adding a layer of enhanced feature extractor and optimizing YOLOV network in cooperation with the previous image preprocessing algorithm, the wafer defect extraction effect can be better.
The network structure diagram shown in fig. 7, compared with the original YOLOV model, adds an up-sampling, which is more obvious for extracting the defect characteristics of the wafer surface.
In a second aspect, the present application provides an intelligent wafer surface defect detection apparatus, including:
The wafer visual photographing device comprises a detection workbench 7, a wafer visual photographing module 2 and a multi-degree-of-freedom robot 6, wherein the wafer visual photographing module 2 and the multi-degree-of-freedom robot 6 are symmetrically arranged on the detection workbench 7, the wafer visual photographing module 2 is used for photographing images of wafer surface defects, and the multi-degree-of-freedom robot 6 is used for driving the wafer 3 or the wafer visual photographing module 2 to perform position movement transformation.
As shown in fig. 6, the wafer vision photographing module 2 includes an industrial camera one 2-1, an industrial camera two 2-3, a high-intensity light source 2-5 and an LED diffuse parallel light source 2-6, wherein the industrial camera one 2-1 and the LED diffuse parallel light source 2-6 are used for collecting the wafer surface image in a detection one mode, and the industrial camera two 2-3 and the high-intensity light source 2-5 are used for collecting the wafer surface image in a detection two mode, wherein the industrial camera one 2-1 and the industrial camera two 2-3 are respectively provided with an industrial lens one 2-2 and an industrial lens two 2-4, and the industrial camera one 2-1 and the industrial camera two 2-3 are fixedly mounted on a sensor bracket 2-7.
The intelligent detection method for the surface defects of the wafer can accurately and efficiently detect various defects on the surface of the wafer in a high-degree-of-freedom polishing mode, and can achieve the aim of detecting the surface defects of the wafer in multiple angles, full sizes and full types through the mutual matching of different detection modes.
In the foregoing embodiments of the present application, the descriptions of the embodiments are emphasized, and for a portion of this disclosure that is not described in detail in this embodiment, reference is made to the related descriptions of other embodiments.
It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media having computer-usable program code embodied therein. The storage medium may be implemented by any type of volatile or non-volatile Memory device or combination thereof, such as static random access Memory (Static Random Access Memory, SRAM), electrically erasable Programmable Read-Only Memory (ELECTRICALLY ERASABLE PROGRAMMABLE READ-Only Memory, EEPROM), erasable Programmable Read-Only Memory (Erasable Programmable Read Only Memory, EPROM), programmable Read-Only Memory (PROM), read-Only Memory (ROM), magnetic Memory, flash Memory, magnetic disk, or optical disk. These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
In the embodiments provided in the present application, it should be understood that the disclosed apparatus and method may be implemented in other manners. The above-described apparatus embodiments are merely illustrative, for example, the division of the units is merely a logical function division, and there may be other manners of division in actual implementation, and for example, multiple units or components may be combined or integrated into another system, or some features may be omitted, or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be through some communication interface, device or unit indirect coupling or communication connection, which may be in electrical, mechanical or other form.
Claims (5)
1. The intelligent detection method for the wafer surface defects is characterized by providing an intelligent detection device for the wafer surface defects, wherein the intelligent detection device for the wafer surface defects comprises a detection workbench, a wafer vision photographing module and a multi-degree-of-freedom robot, the wafer vision photographing module and the multi-degree-of-freedom robot are symmetrically arranged on the detection workbench, the wafer vision photographing module is used for photographing images of the wafer surface defects, and the multi-degree-of-freedom robot is used for driving the wafer or the wafer vision photographing module to perform position movement transformation;
the detection method comprises the following steps:
The wafer and the wafer vision photographing module are respectively arranged on the corresponding carriers;
Controlling the wafer or the wafer vision photographing module to move to a first detection initial relative position;
The wafer vision photographing module is used for collecting the surface image of the wafer in a detection mode, and photographing is carried out on path points by path points under a preset detection path in the detection mode, and circulation is carried out to the end point of the detection path;
Controlling the wafer or the wafer vision photographing module to move to a second detection initial relative position;
the wafer vision photographing module is used for collecting the surface image of the wafer in the two detection modes, including photographing by path points under a preset detection path in the two detection modes, and circulating to the detection path end point;
Performing image detection processing on all shot images obtained in the first detection mode and the second detection mode to generate a wafer surface defect detection result;
The detection of the first mode is carried out by polishing the surface of the wafer through the LED diffuse parallel light source under different inclination angles;
The detection two modes polish the surface of the wafer under different inclination angles through a high-intensity light source with illuminance larger than a first threshold value;
The polishing of the surface of the wafer is carried out under different inclination angles in the first detection mode and the second detection mode, wherein the polishing angle alpha is an included angle between the emergent direction of the light source and the normal direction of the surface of the wafer, the high polishing angle range is alpha=0-30 degrees, and the low polishing angle range is alpha=60-90 degrees;
in a detection mode, when the included angle between the emergent direction of the light source and the normal direction of the wafer surface is alpha=60-90 degrees, the collected images highlight the defects of claw cracks, scratches and scratch, and the contrast ratio of the defect area and the non-defect area is enhanced; when the included angle between the emergent direction of the light source and the normal direction of the wafer surface is alpha=0-30 degrees, the acquired image highlights the contrast of the wafer edge image, and the gray level of the wafer image is more uniform;
In the second detection mode, when the included angle between the outgoing direction of the high-intensity light source and the normal direction of the wafer surface is alpha=60-90 degrees, the collected images highlight contamination and the pinhole defect area show bright spots to be converged, and when the included angle between the outgoing direction of the high-intensity light source and the normal direction of the wafer surface is alpha=0-30 degrees, the area burnt by the liquid medicine residue is developed into a bright surface or shadow under high-intensity light irradiation, so that the defect detection algorithm is easy to extract the bright surface or shadow.
2. The intelligent wafer surface defect detection method according to claim 1, wherein the steps of installing the wafer and the wafer vision photographing module on the corresponding carriers respectively, and controlling the wafer or the wafer vision photographing module to move to the first detection starting relative position comprise adsorbing the wafer to be detected on a sucker on a terminal support of the multi-degree-of-freedom robot, fixedly installing the wafer vision photographing module on the module support, and controlling the multi-degree-of-freedom robot to drive the wafer to be detected to move to the first detection starting relative position.
3. The intelligent wafer surface defect detection method according to claim 1, wherein the steps of installing the wafer and the wafer vision photographing module on the corresponding carriers respectively, and controlling the wafer or the wafer vision photographing module to move to the first detection start relative position comprise installing the wafer vision photographing module on a terminal support of the multi-degree-of-freedom robot, placing the wafer to be detected on the module support, and controlling the multi-degree-of-freedom robot to drive the vision photographing module to move to the first detection start relative position.
4. The intelligent wafer surface defect detection method according to claim 3, wherein the image detection processing is performed on all the shot images obtained in the first mode and the second mode, the wafer surface defect detection result is generated by performing the region segmentation processing on all the shot images after the pretreatment, so as to cut out the wafer image, and the optimized YOLOV model is used for finishing the surface defect target detection of the wafer image.
5. The intelligent wafer surface defect detection method according to claim 4, wherein the wafer vision photographing module comprises an industrial camera I, an industrial camera II, a high-intensity light source and an LED diffuse parallel light source, wherein the industrial camera I and the LED diffuse parallel light source are used for collecting wafer surface images in a detection mode, and the industrial camera II and the high-intensity light source are used for collecting wafer surface images in a detection mode.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310008991.7A CN116183623B (en) | 2023-01-04 | 2023-01-04 | Intelligent wafer surface defect detection method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310008991.7A CN116183623B (en) | 2023-01-04 | 2023-01-04 | Intelligent wafer surface defect detection method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116183623A CN116183623A (en) | 2023-05-30 |
| CN116183623B true CN116183623B (en) | 2024-04-26 |
Family
ID=86443641
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310008991.7A Active CN116183623B (en) | 2023-01-04 | 2023-01-04 | Intelligent wafer surface defect detection method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116183623B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117541531B (en) * | 2023-09-28 | 2024-06-14 | 苏州梅曼智能科技有限公司 | Wafer vision detection supervision feedback system based on artificial intelligence |
| CN117410211B (en) * | 2023-12-11 | 2024-03-22 | 天通控股股份有限公司 | Coding and defect identification system and control method for boxed wafers |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH0763537A (en) * | 1993-08-30 | 1995-03-10 | Nippondenso Co Ltd | Surface defect inspecting device |
| JPH08285782A (en) * | 1995-04-11 | 1996-11-01 | Nikon Corp | Macro image inspection equipment |
| CN101924053A (en) * | 2009-01-13 | 2010-12-22 | 联达科技设备私人有限公司 | System and method for inspecting a wafer |
| KR20160068228A (en) * | 2014-12-05 | 2016-06-15 | 세메스 주식회사 | Apparatus for inspecting defects of a substrate |
| CN111982931A (en) * | 2020-08-27 | 2020-11-24 | 惠州高视科技有限公司 | High-precision wafer surface defect detection device and detection method thereof |
| CN215179575U (en) * | 2021-03-23 | 2021-12-14 | 王唐(苏州)机器人智能科技有限公司 | Device for detecting appearance defects of upper surface of chip |
| CN114723833A (en) * | 2022-04-08 | 2022-07-08 | 广西师范大学 | Improved YOLOV 5-based deep learning wafer solder joint detection method |
| CN115184376A (en) * | 2022-07-29 | 2022-10-14 | 昂坤视觉(北京)科技有限公司 | Wafer detection method and device |
| CN217655027U (en) * | 2022-07-23 | 2022-10-25 | 江阴市华拓芯片测试有限公司 | Quick screening device of target defect on wafer |
| CN115423796A (en) * | 2022-09-22 | 2022-12-02 | 重庆邮电大学 | Chip defect detection method and system based on TensorRT accelerated reasoning |
| CN115439427A (en) * | 2022-08-26 | 2022-12-06 | 常州大学 | Wafer defect detection and positioning algorithm based on cascade YOLO-GAN |
-
2023
- 2023-01-04 CN CN202310008991.7A patent/CN116183623B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH0763537A (en) * | 1993-08-30 | 1995-03-10 | Nippondenso Co Ltd | Surface defect inspecting device |
| JPH08285782A (en) * | 1995-04-11 | 1996-11-01 | Nikon Corp | Macro image inspection equipment |
| CN101924053A (en) * | 2009-01-13 | 2010-12-22 | 联达科技设备私人有限公司 | System and method for inspecting a wafer |
| KR20160068228A (en) * | 2014-12-05 | 2016-06-15 | 세메스 주식회사 | Apparatus for inspecting defects of a substrate |
| CN111982931A (en) * | 2020-08-27 | 2020-11-24 | 惠州高视科技有限公司 | High-precision wafer surface defect detection device and detection method thereof |
| CN215179575U (en) * | 2021-03-23 | 2021-12-14 | 王唐(苏州)机器人智能科技有限公司 | Device for detecting appearance defects of upper surface of chip |
| CN114723833A (en) * | 2022-04-08 | 2022-07-08 | 广西师范大学 | Improved YOLOV 5-based deep learning wafer solder joint detection method |
| CN217655027U (en) * | 2022-07-23 | 2022-10-25 | 江阴市华拓芯片测试有限公司 | Quick screening device of target defect on wafer |
| CN115184376A (en) * | 2022-07-29 | 2022-10-14 | 昂坤视觉(北京)科技有限公司 | Wafer detection method and device |
| CN115439427A (en) * | 2022-08-26 | 2022-12-06 | 常州大学 | Wafer defect detection and positioning algorithm based on cascade YOLO-GAN |
| CN115423796A (en) * | 2022-09-22 | 2022-12-02 | 重庆邮电大学 | Chip defect detection method and system based on TensorRT accelerated reasoning |
Non-Patent Citations (1)
| Title |
|---|
| 韩至成 等.太阳能级硅提纯技术与装备.冶金工业出版社,2011,第294-296页. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116183623A (en) | 2023-05-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116183623B (en) | Intelligent wafer surface defect detection method and device | |
| JP5548287B2 (en) | Offset correction method and apparatus for positioning and inspecting a substrate | |
| US9053390B2 (en) | Automated inspection scenario generation | |
| CN108645867B (en) | Rapid locating and batch detection method for micro defects on surface of large-caliber optical crystal | |
| CN115020267B (en) | Semiconductor surface defect detection method | |
| US10161883B2 (en) | Wafer inspection method and wafer inspection apparatus | |
| US9406115B2 (en) | Scratch detection method and apparatus | |
| WO2006046236A1 (en) | Method and apparatus for residue detection on a polished wafer | |
| CN111474179A (en) | Lens surface cleanliness detection device and method | |
| CN114813783B (en) | Chip surface flaw detection method | |
| CN108422321A (en) | The polishing pad image detecting system and method for chemical mechanical grinding | |
| CN115112682A (en) | Visual mobile phone curved glass detection device and method | |
| CN208514344U (en) | polishing pad image detection system for chemical mechanical polishing | |
| CN114820475A (en) | Edge identification method and system, wafer processing device and method for determining concentric state of wafer and processing table | |
| JP4146678B2 (en) | Image inspection method and apparatus | |
| CN114113112B (en) | Surface micro defect positioning and identifying method based on three-light-source microscopic system | |
| CN117169247A (en) | Metal surface defect multi-dimensional detection method and system based on machine vision | |
| JP2017069579A (en) | Edge detector | |
| JP3632501B2 (en) | Crystal defect inspection method and inspection apparatus | |
| CN210180940U (en) | Automatic optical detection mechanism for detecting defects of silicon wafer | |
| TWM585898U (en) | Automatic optical inspection mechanism for inspecting defects of silicon wafer | |
| CN111583137A (en) | Electronic screen color spot detection device and method based on machine vision technology | |
| TWI691715B (en) | Automatic optical detection mechanism for detecting silicon wafer defects and method thereof | |
| CN116523882B (en) | Vision-based optical target area accuracy detection method and system | |
| Lu et al. | High Precision Image Stabilization Control for Wafer Defect Inspection |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |