Real time recipe tuning for inspection system

Abstract

The present disclosure provides a method for real time parameter tuning for a wafer inspection system. The method comprises acquiring one or more input images streamed from the wafer inspection system; applying a plurality of image enhancement parameters to the one or more input image to generate a plurality of images with different characteristics; identifying defects from the plurality of images by applying a plurality of defect detection parameters; and determining, based on the defects identified on the plurality of images, a parameter combination of the plurality of image enhancement parameters and of the plurality of the defect detection parameters to be used for detecting a first defect type.

Description

REAL TIME RECIPE TUNING FOR INSPECTION SYSTEMCROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims priority of US application 63 / 538,268 which was filed on 13 September 2023, and which is incorporated herein in its entirety by reference.TECHNICAL FIELD

[0002] The embodiments provided herein generally relate to inspection recipe tuning technology, and more particularly, to real time inspection tuning technology utilizing an offline computation platform.BACKGROUND

[0003] In manufacturing processes of integrated circuits (ICs), unfinished or finished circuit components are inspected to ensure that they are manufactured according to design and are free of defects. Inspection systems utilizing optical microscopes or charged particle (e.g., electron) beam microscopes, such as a scanning electron microscope (SEM) can be employed. As the physical sizes of IC components continue to shrink, accuracy and yield in defect detection becomes more and more important. Accordingly, precise controlling of an inspection tool such as SCPM tool is important for accurate defect detection and metrology. Because the process of record (POR) recipe of an inspection tool generally contains thousands of parameters, some of which are interdependent, inspection tool recipe tuning is labor intensive and time consuming, which results in long cycle time. There is a demand in the industry to provide a recipe tuning technology without degrading throughput of the inspection system.SUMMARY

[0004] Some embodiments of the present disclosure can provide a method for real time parameter tuning for a wafer inspection system. The method comprises acquiring one or more input images streamed from the wafer inspection system; applying a plurality of image enhancement parameters to the one or more input images to generate a plurality of images with different characteristics; identifying defects from the plurality of images by applying a plurality of defect detection parameters; and determining, based on the defects identified on the plurality of images, a parameter combination of the plurality of image enhancement parameters and of the plurality of the defect detection parameters to be used for detecting a first defect type.

[0005] Some embodiments of the present disclosure can provide a method for real time parameter tuning for a wafer inspection system. The method comprises acquiring one or more input images streamed from the wafer inspection system; applying a first parameter combination to the one or more input images and identifying a first set of defects of the one or more input images; applying a second parameter combination to the one or more input images and identifying a second set of defects of theone or more input images; determining a first key performance indicator (KPI) value for the first parameter combination based on the first set of defects; determining a second KPI value for the second parameter combination based on the second set of defects; and ranking the first parameter combination and the second parameter combination based on the first KPI value and the second KPI value.

[0006] Some embodiments of the present disclosure can provide an apparatus for real time parameter tuning for a wafer inspection system. The apparatus comprises a memory storing a set of instructions; and at least one processor configured to execute the set of instructions to cause the apparatus to perform operations comprising: acquiring one or more input images streamed from the wafer inspection system; applying a plurality of image enhancement parameters to the one or more input images to generate a plurality of images with different characteristics; identifying defects from the plurality of images by applying a plurality of defect detection parameters; and determining, based on the defects identified on the plurality of images, a parameter combination of the plurality of image enhancement parameters and of the plurality of the defect detection parameters to be used for detecting a first defect type.

[0007] Other advantages of the embodiments of the present disclosure will become apparent from the following description taken in conjunction with the accompanying drawings wherein are set forth, by way of illustration and example, certain embodiments of the present invention.BRIEF DESCRIPTION OF FIGURES

[0008] The above and other aspects of the present disclosure will become more apparent from the description of exemplary embodiments, taken in conjunction with the accompanying drawings.

[0009] FIG. 1 is a schematic diagram illustrating an example charged-particle beam inspection system, consistent with embodiments of the present disclosure.

[0010] FIG. 2A is a schematic diagram illustrating an example multi-beam tool, consistent with embodiments of the present disclosure that can be a part of the exemplary charged-particle beam inspection system of FIG. 1.

[0011] FIG. 2B is a schematic diagram illustrating an example single-beam tool, consistent with embodiments of the present disclosure that can be a part of the exemplary charged-particle beam inspection system of FIG. 1.

[0012] FIG. 3 is a flow diagram of an example real time recipe tuning method for an inspection tool, consistent with embodiments of the present disclosure.

[0013] FIG. 4A is an example parameter structure for parameter setup, consistent with embodiments of the present disclosure.

[0014] FIG. 4B illustrates example results after applying various image enhancement parameters to input images, consistent with embodiments of the present disclosure.

[0015] FIG. 4C illustrates example resultant images after applying various image enhancement parameters to an input image, consistent with embodiments of the present disclosure.

[0016] FIG. 5A illustrates example parameter combination results after applying various detection parameters to enhanced images, consistent with embodiments of the present disclosure.

[0017] FIG. 5B illustrates an example resultant image that is suitable for critical dimension measurement, consistent with embodiments of the present disclosure.

[0018] FIG. 5C illustrates an example resultant image that is suitable for bridge defect type detection, consistent with embodiments of the present disclosure.

[0019] FIG. 5D illustrates an example resultant image that is suitable for broken defect type detection, consistent with embodiments of the present disclosure.

[0020] FIG. 6A illustrates example parameter combination results after applying various threshold parameters for a first defect type, consistent with embodiments of the present disclosure.

[0021] FIG. 6B illustrates effects of a threshold parameter change in detecting a defect, consistent with embodiments of the present disclosure.

[0022] FIG. 7A illustrates a first example analysis result for a set parameter combination, consistent with embodiments of the present disclosure.

[0023] FIG. 7B illustrates a second example analysis result for a set parameter combination, consistent with embodiments of the present disclosure.

[0024] FIG. 7C illustrates an example KPI map per parameter combination, consistent with embodiments of the present disclosure.

[0025] FIG. 8 is a flow chart illustrating an example method for real time recipe tuning of an inline inspection system, consistent with embodiments of the present disclosure.

[0026] FIG. 9 is a block diagram of an example computer system, consistent with embodiments of the present disclosure.DETAILED DESCRIPTION

[0027] Reference will now be made in detail to exemplary embodiments, examples of which are illustrated in the accompanying drawings. The following description refers to the accompanying drawings in which the same numbers in different drawings represent the same or similar elements unless otherwise represented. The implementations set forth in the following description of exemplary embodiments do not represent all implementations. Instead, they are merely examples of apparatuses and methods consistent with aspects related to the disclosed embodiments as recited in the appended claims. For example, although some embodiments are described in the context of utilizing electron beams, the disclosure is not so limited. Other types of charged particle beams (e.g., including protons, ions, muons, or any other particle carrying electric charges) may be similarly applied. Furthermore, other imaging systems may be used, such as optical imaging, photon detection, x-ray detection, ion detection, etc.

[0028] Electronic devices are constructed of circuits formed on a piece of semiconductor material called a substrate. The semiconductor material may include, for example, silicon, gallium arsenide, indium phosphide, or silicon germanium, or the like. Many circuits may be formed together on the same piece of silicon and are called integrated circuits or ICs. The size of these circuits has decreased dramatically so that many more of them can be fit on the substrate. For example, an IC chip in a smartphone can be as small as a thumbnail and yet may include over 2 billion transistors, the size of each transistor being less than l / 1000th the size of a human hair.

[0029] Making these ICs with extremely small structures or components is a complex, timeconsuming, and expensive process, often involving hundreds of individual steps. Errors in even one step have the potential to result in defects in the finished IC, rendering it useless. Thus, one goal of the manufacturing process is to avoid such defects to maximize the number of functional ICs made in the process; that is, to improve the overall yield of the process.

[0030] One component of improving yield is monitoring the chip-making process to ensure that it is producing a sufficient number of functional integrated circuits. One way to monitor the process is to inspect the chip circuit structures at various stages of their formation. Inspection can be carried out using a scanning charged-particle microscope (“SCPM”). For example, an SCPM may be a scanning electron microscope (SEM). A SCPM can be used to image these extremely small structures, in effect, taking a “picture” of the structures of the wafer. The image can be used to determine if the structure was formed properly in the proper location. If the structure is defective, then the process can be adjusted, so the defect is less likely to recur.

[0031] As the physical sizes of IC components continue to shrink, accuracy and yield in defect detection becomes more and more important. Accordingly, precise controlling of an inspection tool such as a SCPM tool is important for accurate defect detection and metrology. Because the process of record (POR) recipe of an inspection tool generally contains thousands of parameters, some of which are interdependent, tuning the inspection tool recipe is labor intensive and time consuming, which results in long cycle time. The current recipe tuning method highly depends on user’s experience, and thus the recipe tuning performance may not be consistent and the resultant recipe could be not robust. Further, the current inspection tool used for inline inspection has limited computational resources and is not possible to support additional computationally intensive processing such as recipe tuning because image processing for defect detection and metrology has already been the bottle neck in the high-volume manufacturing (HVM) wafer process. If the inspection tool’s computational resources are assigned for recipe tuning in the middle of the production, it will consume wafer cycle time and thus degrade throughput of the HVM wafer process. As discussed, there is a trade-off between inspection sensitivity and throughput for wafer cycle time. Therefore, the HVM production recipe is generally set to capture specific defect types under certain detection thresholds and image enhancement conditions and is maintained throughout the production. However, the HVM wafer process tends to experience process variations as time passes, which will result in missing defects dueto the fixed production recipe. Moreover, because the HVM process is set to use the same threshold for a certain defect type during entire inspection of all HVM wafers, it may not be possible to capture new defects or new defect types without having to increase wafer cycle time once the production begins. There is a demand in the industry to provide a recipe tuning method without degrading throughput of the inspection system.

[0032] According to some embodiments of the present disclosure, a real time recipe tuning technique for an inspection system is provided. According to some embodiments of the present disclosure, an offline computation platform, which is separate from an inline inspection tool used for production, is utilized for a real time recipe tuning for the inline inspection tool. According to some embodiments of the present disclosure, comprehensive and computationally intensive recipe tuning for image inspection or metrology can be performed in parallel with the inline production without increasing wafer cycle time. According to some embodiments of the present disclosure, a parameter combination that can improve a defect capture rate including a new defect detection or a new defect type detection can be provided. According to some embodiments of the present disclosure, a weak point for subsequent inspection can be predicted.

[0033] Relative dimensions of components in drawings may be exaggerated for clarity. Within the following description of drawings, the same or like reference numbers refer to the same or like components or entities, and only the differences with respect to the individual embodiments are described. Other objects and advantages of the disclosure may be realized by the elements and combinations as set forth in the embodiments discussed herein. However, embodiments of the present disclosure are not necessarily required to achieve such exemplary objects or advantages, and some embodiments may not achieve any of the stated objects or advantages.

[0034] Without limiting the scope of the present disclosure, some embodiments may be described in the context of providing scanning deflection systems and scanning deflection methods in systems utilizing electron beams (“e-beams”). Some scanning deflection systems may use electric fields to influence a charged particle beam. However, the disclosure is not so limited. Other types of charged particle beams may be similarly applied. For example, systems and methods may be applicable with optics, photons, x-rays, and ions, etc. Deflection may be used to scan a beam over a surface in, for example, cathode ray tubes (CRTs), lithography machines, scanning charged-particle microscopes (SCPMs), or other analytical instruments. While some embodiments are discussed with reference to deflection systems that use electric field to influence a beam, deflection may also be achieved with magnetic fields, for example.

[0035] As used herein, unless specifically stated otherwise, the term “or” encompasses all possible combinations, except where infeasible. For example, if it is stated that a component includes A or B, then, unless specifically stated otherwise or infeasible, the component may include A, or B, or A and B. As a second example, if it is stated that a component includes A, B, or C, then, unless specifically stated otherwise or infeasible, the component may include A, or B, or C, or A and B, or A and C, or Band C, or A and B and C. Expressions such as “at least one of’ do not necessarily modify an entirety of a following list and do not necessarily modify each member of the list, such that “at least one of A, B, and C” should be understood as including only one of A, only one of B, only one of C, or any combination of A, B, and C. The phrase “one of A and B” or “any one of A and B” shall be interpreted in the broadest sense to include one of A, or one of B.

[0036] FIG. 1 illustrates an example electron beam inspection (EBI) system 100 consistent with embodiments of the present disclosure. EBI system 100 may be used for imaging. As shown in FIG. 1, EBI system 100 includes a main chamber 101, a load / lock chamber 102, a beam tool 104, and an equipment front end module (EFEM) 106. Beam tool 104 is located within main chamber 101.EFEM 106 includes a first loading port 106a and a second loading port 106b. EFEM 106 may include additional loading port(s). First loading port 106a and second loading port 106b receive wafer front opening unified pods (FOUPs) that contain wafers (e.g., semiconductor wafers or wafers made of other material(s)) or samples to be inspected (wafers and samples may be used interchangeably). A “lot” is a plurality of wafers that may be loaded for processing as a batch.

[0037] One or more robotic arms (not shown) in EFEM 106 may transport the wafers to load / lock chamber 102. Load / lock chamber 102 is connected to a load / lock vacuum pump system (not shown) which removes gas molecules in load / lock chamber 102 to reach a first pressure below the atmospheric pressure. After reaching the first pressure, one or more robotic arms (not shown) may transport the wafer from load / lock chamber 102 to main chamber 101. Main chamber 101 is connected to a main chamber vacuum pump system (not shown) which removes gas molecules in main chamber 101 to reach a second pressure below the first pressure. After reaching the second pressure, the wafer is subject to inspection by beam tool 104. Beam tool 104 may be a single -beam system or a multi-beam system.

[0038] A controller 109 is electronically connected to beam tool 104. Controller 109 may be a computer configured to execute various controls of EBI system 100. While controller 109 is shown in FIG. 1 as being outside of the structure that includes main chamber 101, load / lock chamber 102, and EFEM 106, it is appreciated that controller 109 may be a part of the structure.

[0039] In some embodiments, controller 109 may include one or more processors (not shown). A processor may be a generic or specific electronic device capable of manipulating or processing information. For example, the processor may include any combination of any number of a central processing unit (or “CPU”), a graphics processing unit (or “GPU”), an optical processor, a programmable logic controllers, a microcontroller, a microprocessor, a digital signal processor, an intellectual property (IP) core, a Programmable Logic Array (PL A), a Programmable Array Logic (PAL), a Generic Array Logic (GAL), a Complex Programmable Logic Device (CPLD), a Field- Programmable Gate Array (FPGA), a System On Chip (SoC), an Application-Specific Integrated Circuit (ASIC), and any type circuit capable of data processing. The processor may also be a virtualprocessor that includes one or more processors distributed across multiple machines or devices coupled via a network.

[0040] In some embodiments, controller 109 may further include one or more memories (not shown). A memory may be a generic or specific electronic device capable of storing codes and data accessible by the processor (e.g., via a bus). For example, the memory may include any combination of any number of a random-access memory (RAM), a read-only memory (ROM), an optical disc, a magnetic disk, a hard drive, a solid-state drive, a flash drive, a security digital (SD) card, a memory stick, a compact flash (CF) card, or any type of storage device. The codes and data may include an operating system (OS) and one or more application programs (or “apps”) for specific tasks. The memory may also be a virtual memory that includes one or more memories distributed across multiple machines or devices coupled via a network.

[0041] FIG. 2A illustrates a schematic diagram of an example multi-beam beam tool 104A (also referred to herein as apparatus 104 A) and an image processing system 290 that may be configured for use in EBI system 100 (FIG. 1), consistent with embodiments of the present disclosure.

[0042] Beam tool 104A comprises a charged-particle source 202, a gun aperture 204, a condenser lens 206, a primary charged-particle beam 210 emitted from charged-particle source 202, a source conversion unit 212, a plurality of beamlets 214, 216, and 218 of primary charged-particle beam 210, a primary projection optical system 220, a motorized wafer stage 280, a wafer holder 282, multiple secondary charged-particle beams 236, 238, and 240, a secondary optical system 242, and a charged- particle detection device 244. Primary projection optical system 220 can comprise a beam separator 222, a deflection scanning unit 226, and an objective lens 228. Charged-particle detection device 244 can comprise detection sub-regions 246, 248, and 250.

[0043] Charged-particle source 202, gun aperture 204, condenser lens 206, source conversion unit 212, beam separator 222, deflection scanning unit 226, and objective lens 228 can be aligned with a primary optical axis 260 of apparatus 104A. Secondary optical system 242 and charged-particle detection device 244 can be aligned with a secondary optical axis 252 of apparatus 104A.

[0044] Charged-particle source 202 can emit one or more charged particles, such as electrons, protons, ions, muons, or any other particle carrying electric charges. In some embodiments, charged- particle source 202 may be an electron source. For example, charged-particle source 202 may include a cathode, an extractor, or an anode, wherein primary electrons can be emitted from the cathode and extracted or accelerated to form primary charged-particle beam 210 (in this case, a primary electron beam) with a crossover (virtual or real) 208. For ease of explanation without causing ambiguity, electrons are used as examples in some of the descriptions herein. However, it should be noted that any charged particle may be used in any embodiment of this disclosure, not limited to electrons. Primary charged-particle beam 210 can be visualized as being emitted from crossover 208. Gun aperture 204 can block off peripheral charged particles of primary charged-particle beam 210 to reduce Coulomb effect. The Coulomb effect may cause an increase in size of probe spots.

[0045] Source conversion unit 212 can comprise an array of image-forming elements and an array of beam-limit apertures. The array of image-forming elements can comprise an array of microdeflectors or micro-lenses. The array of image-forming elements can form a plurality of parallel images (virtual or real) of crossover 208 with a plurality of beamlets 214, 216, and 218 of primary charged-particle beam 210. The array of beam-limit apertures can limit the plurality of beamlets 214, 216, and 218. While three beamlets 214, 216, and 218 are shown in FIG. 2A, embodiments of the present disclosure are not so limited. For example, in some embodiments, the apparatus 104A may be configured to generate a first number of beamlets. In some embodiments, the first number of beamlets may be in a range from 1 to 1000. In some embodiments, the first number of beamlets may be in a range from 200-500. In some embodiments, an apparatus 104A may generate 400 beamlets.

[0046] Condenser lens 206 can focus primary charged-particle beam 210. The electric currents of beamlets 214, 216, and 218 downstream of source conversion unit 212 can be varied by adjusting the focusing power of condenser lens 206 or by changing the radial sizes of the corresponding beam-limit apertures within the array of beam-limit apertures. Objective lens 228 can focus beamlets 214, 216, and 218 onto a wafer 230 for imaging, and can form a plurality of probe spots 270, 272, and 274 on a surface of wafer 230.

[0047] Beam separator 222 can be a beam separator of Wien filter type generating an electrostatic dipole field and a magnetic dipole field. In some embodiments, if they are applied, the force exerted by the electrostatic dipole field on a charged particle (e.g., an electron) of beamlets 214, 216, and 218 can be substantially equal in magnitude and opposite in a direction to the force exerted on the charged particle by magnetic dipole field. Beamlets 214, 216, and 218 can, therefore, pass straight through beam separator 222 with zero deflection angle. However, the total dispersion of beamlets 214, 216, and 218 generated by beam separator 222 can also be non-zero. Beam separator 222 can separate secondary charged-particle beams 236, 238, and 240 from beamlets 214, 216, and 218 and direct secondary charged-particle beams 236, 238, and 240 towards secondary optical system 242.

[0048] Deflection scanning unit 226 can deflect beamlets 214, 216, and 218 to scan probe spots 270, 272, and 274 over a surface area of wafer 230. In response to the incidence of beamlets 214, 216, and 218 at probe spots 270, 272, and 274, secondary charged-particle beams 236, 238, and 240 may be emitted from wafer 230. Secondary charged-particle beams 236, 238, and 240 may comprise charged particles (e.g., electrons) with a distribution of energies. For example, secondary charged-particle beams 236, 238, and 240 may be secondary electron beams including secondary electrons (energies < 50 eV) and backscattered electons (energies between 50 eV and landing energies of beamlets 214, 216, and 218). Secondary optical system 242 can focus secondary charged-particle beams 236, 238, and 240 onto detection sub-regions 246, 248, and 250 of charged-particle detection device 244. Detection sub-regions 246, 248, and 250 may be configured to detect corresponding secondary charged-particle beams 236, 238, and 240 and generate corresponding signals (e.g., voltage, current,or the like) used to reconstruct an inspection image of structures on or underneath the surface area of wafer 230.

[0049] The generated signals may represent intensities of secondary charged-particle beams 236, 238, and 240 and may be provided to image processing system 290 that is in communication with charged-particle detection device 244, primary projection optical system 220, and motorized wafer stage 280. The movement speed of motorized wafer stage 280 may be synchronized and coordinated with the beam deflections controlled by deflection scanning unit 226, such that the movement of the scan probe spots (e.g., scan probe spots 270, 272, and 274) may orderly cover regions of interest on the wafer 230. The parameters of such synchronization and coordination may be adjusted to adapt to different materials of wafer 230. For example, different materials of wafer 230 may have different resistance-capacitance characteristics that may cause different signal sensitivities to the movement of the scan probe spots.

[0050] The intensity of secondary charged-particle beams 236, 238, and 240 may vary according to the external or internal structure of wafer 230, and thus may indicate whether wafer 230 includes defects. Moreover, as discussed above, beamlets 214, 216, and 218 may be projected onto different locations of the top surface of wafer 230, or different sides of local structures of wafer 230, to generate secondary charged-particle beams 236, 238, and 240 that may have different intensities. Therefore, by mapping the intensity of secondary charged-particle beams 236, 238, and 240 with the areas of wafer 230, image processing system 290 may reconstruct an image that reflects the characteristics of internal or external structures of wafer 230.

[0051] In some embodiments, image processing system 290 may include an image acquirer 292, a storage 294, and a controller 296. Image acquirer 292 may comprise one or more processors. For example, image acquirer 292 may comprise a computer, server, mainframe host, terminals, personal computer, any kind of mobile computing devices, or the like, or a combination thereof. Image acquirer 292 may be communicatively coupled to charged-particle detection device 244 of beam tool 104A through a medium such as an electric conductor, optical fiber cable, portable storage media, IR, Bluetooth, internet, wireless network, wireless radio, or a combination thereof. In some embodiments, image acquirer 292 may receive a signal from charged-particle detection device 244 and may construct an image. Image acquirer 292 may thus acquire inspection images of wafer 230. Image acquirer 292 may also perform various post-processing functions, such as generating contours, superimposing indicators on an acquired image, or the like. Image acquirer 292 may be configured to perform adjustments of brightness and contrast of acquired images. In some embodiments, storage 294 may be a storage medium such as a hard disk, flash drive, cloud storage, random access memory (RAM), other types of computer-readable memory, or the like. Storage 294 may be coupled with image acquirer 292 and may be used for saving scanned raw image data as original images, and postprocessed images. Image acquirer 292 and storage 294 may be connected to controller 296. In someembodiments, image acquirer 292, storage 294, and controller 296 may be integrated together as one control unit.

[0052] In some embodiments, image acquirer 292 may acquire one or more inspection images of a wafer based on an imaging signal received from charged-particle detection device 244. An imaging signal may correspond to a scanning operation for conducting charged particle imaging. An acquired image may be a single image comprising a plurality of imaging areas. The single image may be stored in storage 294. The single image may be an original image that may be divided into a plurality of regions. Each of the regions may comprise one imaging area containing a feature of wafer 230. The acquired images may comprise multiple images of a single imaging area of wafer 230 sampled multiple times over a time sequence. The multiple images may be stored in storage 294. In some embodiments, image processing system 290 may be configured to perform image processing steps with the multiple images of the same location of wafer 230. In some embodiments, image processing system 290 may be configured to detect defects on the images or enhanced images according to the POR recipe. In some embodiments, the POR recipe may set defect types for detection and thresholds for the set defect types. In some embodiments, the POR recipe may set image enhancement parameter(s) to be applied to the images acquired by image acquirer 292. In some embodiments, the detected defects along with corresponding locations on a wafer or a reticle can be stored in storage 294.

[0053] In some embodiments, image processing system 290 may include measurement circuits (e.g., analog-to-digital converters) to obtain a distribution of the detected secondary charged particles (e.g., secondary electrons). The charged-particle distribution data collected during a detection time window, in combination with corresponding scan path data of beamlets 214, 216, and 218 incident on the wafer surface, can be used to reconstruct images of the wafer structures under inspection. The reconstructed images can be used to reveal various features of the internal or external structures of wafer 230, and thereby can be used to reveal any defects that may exist in the wafer.

[0054] In some embodiments, the charged particles may be electrons. When electrons of primary charged-particle beam 210 are projected onto a surface of wafer 230 (e.g., probe spots 270, 272, and 274), the electrons of primary charged-particle beam 210 may penetrate the surface of wafer 230 for a certain depth, interacting with particles of wafer 230. Some electrons of primary charged-particle beam 210 may elastically interact with (e.g., in the form of elastic scattering or collision) the materials of wafer 230 and may be reflected or recoiled out of the surface of wafer 230. An elastic interaction conserves the total kinetic energies of the bodies (e.g., electrons of primary charged-particle beam 210) of the interaction, in which the kinetic energy of the interacting bodies does not convert to other forms of energy (e.g., heat, electromagnetic energy, or the like). Such reflected electrons generated from elastic interaction may be referred to as backscattered electrons (BSEs). Some electrons of primary charged-particle beam 210 may inelastically interact with (e.g., in the form of inelastic scattering or collision) the materials of wafer 230. An inelastic interaction does not conserve the totalkinetic energies of the bodies of the interaction, in which some or all of the kinetic energy of the interacting bodies convert to other forms of energy. For example, through the inelastic interaction, the kinetic energy of some electrons of primary charged-particle beam 210 may cause electron excitation and transition of atoms of the materials. Such inelastic interaction may also generate electrons exiting the surface of wafer 230, which may be referred to as secondary electrons (SEs). Yield or emission rates of BSEs and SEs depend on, e.g., the material under inspection and the landing energy of the electrons of primary charged-particle beam 210 landing on the surface of the material, among others. The energy of the electrons of primary charged-particle beam 210 may be imparted in part by its acceleration voltage (e.g., the acceleration voltage between the anode and cathode of charged-particle source 202 in FIG. 2A). The quantity of BSEs and SEs may be more or fewer (or even the same) than the injected electrons of primary charged-particle beam 210.

[0055] Another example of a charged particle beam apparatus will now be discussed with reference to FIG. 2B. Beam tool 104B (also referred to herein as apparatus 104B) may be an example of beam tool 104 and may be similar to beam tool 104A shown in FIG. 2A. However, different from apparatus 104A, apparatus 104B may be a single-beam tool that uses only one primary electron beam to scan one location on the wafer at a time.

[0056] As shown in FIG. 2B, apparatus 104B includes a wafer holder 136 supported by motorized stage 134 to hold a wafer 150 to be inspected. Beam tool 104B includes an electron emitter, which may comprise a cathode 103, an anode 121, and a gun aperture 122. Beam tool 104B further includes a beam limit aperture 125, a condenser lens 126, a column aperture 135, an objective lens assembly 132, and a detector 144. Objective lens assembly 132, in some embodiments, may be a modified SORIL lens, which includes a pole piece 132a, a control electrode 132b, a deflector unit 132c, and an exciting coil 132d. In a detection or imaging process, an electron beam 161 emanating from the tip of cathode 103 may be accelerated by anode 121 voltage, pass through gun aperture 122, beam limit aperture 125, condenser lens 126, and be focused into a probe spot 170 by the modified SORIL lens and impinge onto the surface of wafer 150. Probe spot 170 may be scanned across the surface of wafer 150 by a deflector, such as deflector unit 132c or other deflectors in the SORIL lens. Secondary or scattered particles, such as secondary electrons or scattered primary electrons emanated from the wafer surface may be collected by detector 144 to determine intensity of the beam and so that an image of an area of interest on wafer 150 may be reconstructed.

[0057] There may also be provided an image processing system 199 that includes an image acquirer 120, a storage 130, and controller 109. Image acquirer 120 may comprise one or more processors. For example, image acquirer 120 may comprise a computer, server, mainframe host, terminals, personal computer, any kind of mobile computing devices, and the like, or a combination thereof. Image acquirer 120 may connect with detector 144 of beam tool 104B through a medium such as an electrical conductor, optical fiber cable, portable storage media, IR, Bluetooth, internet, wireless network, wireless radio, or a combination thereof. Image acquirer 120 may receive a signal fromdetector 144 and may construct an image. Image acquirer 120 may thus acquire images of wafer 150. Image acquirer 120 may also perform various post-processing functions, such as image averaging, generating contours, superimposing indicators on an acquired image, and the like. Image acquirer 120 may be configured to perform adjustments of brightness and contrast, etc. of acquired images.Storage 130 may be a storage medium such as a hard disk, random access memory (RAM), cloud storage, other types of computer readable memory, and the like. Storage 130 may be coupled with image acquirer 120 and may be used for saving scanned raw image data as original images, and postprocessed images. Image acquirer 120 and storage 130 may be connected to controller 109. In some embodiments, image acquirer 120, storage 130, and controller 109 may be integrated together as one electronic control unit. In some embodiments, image processing system 199 may be configured to detect defects on the images or enhanced images according to the POR recipe. In some embodiments, the POR recipe may set defect types for detection and thresholds for the set defect types. In some embodiments, the POR recipe may set image enhancement parameter(s) to be applied to the images acquired by image acquirer 120. In some embodiments, the detected defects along with corresponding locations on a wafer or a reticle can be stored in storage 130.

[0058] In some embodiments, image acquirer 120 may acquire one or more images of a sample based on an imaging signal received from detector 144. An imaging signal may correspond to a scanning operation for conducting charged particle imaging. An acquired image may be a single image comprising a plurality of imaging areas that may contain various features of wafer 150. The single image may be stored in storage 130. Imaging may be performed on the basis of imaging frames.

[0059] The condenser and illumination optics of the electron beam tool may comprise or be supplemented by electromagnetic quadrupole electron lenses. For example, as shown in FIG. 2B, electron beam tool 104B may comprise a first quadrupole lens 148 and a second quadrupole lens 158. In some embodiments, the quadrupole lenses may be used for controlling the electron beam. For example, first quadrupole lens 148 may be controlled to adjust the beam current and second quadrupole lens 158 may be controlled to adjust the beam spot size and beam shape.

[0060] FIG. 2B illustrates a charged particle beam apparatus that may use a single primary beam configured to generate secondary electons by interacting with wafer 150. Detector 144 may be placed along optical axis 105, as in the embodiment shown in FIG. 2B. The primary electron beam may be configured to travel along optical axis 105. Accordingly, detector 144 may include a hole at its center so that the primary electron beam may pass through to reach wafer 150. FIG. 2B shows an example of detector 144 having an opening at its center. However, some embodiments may use a detector placed off-axis relative to the optical axis along which the primary electron beam travels. For example, as in the embodiment shown in FIG. 2B, discussed above, a beam separator 222 may be provided to direct secondary electron beams toward a detector placed off-axis. Beam separator 222may be configured to divert secondary electron beams toward an electron detection device 244, as shown in FIG. 2A.

[0061] The images generated by SCPM may be used for defect inspection. For example, a generated image capturing a test device region of a wafer may be compared with a reference image capturing the same test device region. The reference image may be predetermined (e.g., by simulation) and include no known defect. If a difference between the generated image and the reference image exceeds a tolerance level, a potential defect may be identified. For another example, the SCPM may scan multiple regions of the wafer, each region including a test device region designed as the same, and generate multiple images capturing those test device regions as manufactured. The multiple images may be compared with each other. If a difference between the multiple images exceeds a tolerance level, a potential defect may be identified.

[0062] In some embodiments, machine learning may be employed in the generation of inspection images, reference images or other images associated with apparatus 100, 104A, or 104B. For example, in some embodiments a machine learning system may be operated in association with, e.g., controller 109 or 296, image processing system 199 or 290, image acquirer 120 or 292, or storage unit 130 or 294 of FIGs. 1-2B. In some embodiments, machine learning may be employed in the real time recipe tuning method, e.g., method 300 of FIG. 3 or 800 of FIG. 8, in association with, e.g., an offline computation platform 301 of FIG. 3, which will be described below. In some embodiments, a machine learning system may comprise a discriminative model. In some embodiments, a machine learning system may include a generative model. For example, learning can feature two types of mechanisms: discriminative learning that may be used to create classification and detection algorithms, and generative learning that may be used to actually create models that, in the extreme, can render images. For example, as described further below, a generative model may be configured for generating an image from a design clip that resembles a corresponding location on a wafer in a SEM image. This may be performed by 1) training the generative model with design clips and the associated actual SEM images from those locations on the wafer; and 2) using the model in inference mode to feed the model design clips in locations for which simulated SEM images are desired. Such simulated images can be used as reference images in, e.g., die-to-database inspection.

[0063] If the model(s) include one or more discriminative models, the discriminative model(s) may have any suitable architecture or configuration known in the art. Discriminative models, also called conditional models, are a class of models used in machine learning for modeling the dependence of an unobserved variable “y” on an observed variable “x.” Within a probabilistic framework, this may be done by modeling a conditional probability distribution P(ylx), which can be used for predicting y based on x. Discriminative models, as opposed to generative models, may not allow one to generate samples from the joint distribution of x and y. However, for tasks such as classification and regression that do not require the joint distribution, discriminative models may yield superior performance. On the other hand, generative models are typically more flexible than discriminativemodels in expressing dependencies in complex learning tasks. In addition, most discriminative models are inherently supervised and cannot easily be extended to unsupervised learning. Application specific details ultimately dictate the suitability of selecting a discriminative versus generative model.

[0064] A generative model can be generally defined as a model that is probabilistic in nature. In other words, a “generative” model is not one that performs forward simulation or rule -based approaches and, as such, it may not be necessary to model the physics of the processes involved in generating an actual image or output (for which a simulated image or output is being generated). Instead, the generative model can be learned (in that its parameters can be learned) based on a suitable training set of data. Such generative models may have a number of advantages for the embodiments described herein. In addition, the generative model may be configured to have a deep learning architecture in that the generative model may include multiple layers, which may perform a number of algorithms or transformations. The number of layers included in the generative model may depend on the particular use case. For practical purposes, a suitable range of layers is from 2 layers to a few tens of layers.

[0065] Deep learning is a type of machine learning. Machine learning can be generally defined as a type of artificial intelligence (Al) that provides computers with the ability to learn without being explicitly programmed. Machine learning focuses on the development of computer programs that can teach themselves to grow and change when exposed to new data. Machine learning explores the study and construction of algorithms that can learn from and make predictions on data — such algorithms overcome following strictly static program instructions by making data driven predictions or decisions, through building a model from sample inputs.

[0066] The machine learning described herein may be further performed as described in “Introduction to Statistical Machine Learning,” by Sugiyama, Morgan Kaufmann, 2016, 534 pages; “Discriminative, Generative, and Imitative Learning,” Jebara, MIT Thesis, 2002, 212 pages; and “Principles of Data Mining (Adaptive Computation and Machine Learning)” Hand et al., MIT Press, 2001, 578 pages; which are incorporated by reference as if fully set forth herein. The embodiments described herein may be further configured as described in these references.

[0067] In some embodiments, a machine learning system may comprise a neural network. For example, a model may be a deep neural network with a set of weights that model the world according to the data that it has been fed to train it. Neural networks can be generally defined as a computational approach which is based on a relatively large collection of neural units loosely modeling the way a biological brain solves problems with relatively large clusters of biological neurons connected by axons. Each neural unit is connected with many others, and links can be enforcing or inhibitory in their effect on the activation state of connected neural units. These systems are self-learning and trained rather than explicitly programmed and excel in areas where the solution or feature detection is difficult to express in a traditional computer program.

[0068] Neural networks typically consist of multiple layers, and the signal path traverses from front to back. The goal of the neural network is to solve problems in the same way that the human brain would, although several neural networks are much more abstract. Modern neural network projects typically work with a few thousand to a few million neural units and millions of connections. The neural network may have any suitable architecture or configuration known in the art.

[0069] In a further embodiment, a model may comprise convolutional and deconvolution neural network. For example, the embodiments described herein can take advantage of learning concepts such as a convolution and deconvolution neural network to solve the normally intractable representation conversion problem (e.g., rendering). The model may have any convolution and deconvolution neural network configuration or architecture known in the art.

[0070] Reference is now made to FIG. 3, which is a flow diagram of an example real time recipe tuning method 300 for an inspection tool, consistent with embodiments of the present disclosure. According to some embodiments of the present disclosure, real time recipe tuning method 300 can be performed on an offline computation platform 301, which is separate from inline tool 1001. In some embodiments, inline tool 1001 can be an inspection tool, e.g., inspection system 100 of FIG. 1, multibeam tool 104A of FIG. 2A, or single-beam tool 104B of FIG. 2B. In some embodiments, inline tool 1001 can be one or more inline systems that are utilized for the HVM wafer process. For example, when same design data is used by multiple inspection systems in the HVM process, offline computation platform 301 can perform real time recipe tuning method 300 for these multiple inspection systems. In some embodiments, offline computation platform 301 has capability to carry out image processing and defect detection to mimic inspection and metrology performed in the inline tool 1001. Because inline tool 1001 may continue performing its operations while offline computation platform 301 performs recipe tuning method 300, recipe tuning for inline tool 1001 can be performed without any throughput impact on inline tool 1001.

[0071] According to some embodiments of the present disclosure, real time recipe tuning method 300 may begin by acquiring one or more input images 3001 as shown in FIG. 3. In some embodiments, input image 3001 can be acquired from inline tool 1001. In some embodiments, input image 3001 is streamed from inline tool 1001 in real time as it is taken from inline tool 1001. While it is not illustrated in FIG. 3, input image 3001 can be saved on offline computation platform 301, such as a database on offline computation platform 301. According to some embodiments of the present disclosure, input image 3001 can be an inspection image, e.g., generated by EBI system 100 of FIG. 1, multi-beam tool 104A of FIG. 2A, or single-beam tool 104B of FIG. 2B. In some embodiments, an inspection image is a SCPM image of a sample or a wafer. In some embodiments, input image 3001 can be a scanned raw image by the inspection tool. While input image 3001 is in a singular form, it will be appreciated that input image 3001 can refer to a group of input images streamed from inline tool 1001.

[0072] According to some embodiments of the present disclosure, when streaming input image 3001 to offline computation platform 301, inline tool 1001 may stream the POR recipe including parameter information used for generating input image 3001. The POR recipe can also be stored in offline computation platform 301. In some embodiments, inline tool 1001 can also stream its image inspection and metrology results for input image 3001 along with the POR parameter recipe for performing the image inspection and metrology on input image 3001. For example, the POR parameter recipe for performing the image inspection and metrology may comprise an image enhancement parameter, a defect type parameter, a threshold for determining the defect type, etc. In some embodiments where multiple beams are utilized in inline tool 1001, inline tool 1001 may also stream information which beam among the multiple beams is associated with input image 3001. In some embodiment where multiple inline systems are utilized, inline tool 1001 may also stream information which inline system among the multiple inline systems is associated with input image 3001.

[0073] According to some embodiments of the present disclosure, a parameter recipe can be set up at a Design of Experiment (DOE) process 310 for experiments. During DOE process 310, parameter values for a set of parameters of interest can be setup for experiments. In some embodiments, a set of parameters can include at least one parameter that is utilized in inline tool 1001 for inspection and metrology of an inspection image. In some embodiments, DOE process 310 can be iterated to sweep through various combinations of parameters of interest. In some embodiments, experiments with predesigned or pre-determined recipe combination(s) can be automatically triggered as input image 3001 is streamed to offline computation platform 301. According to some embodiments of the present disclosure, input image 3001 is used as input for image inspection for defect detection or metrology with the pre-determined recipe combination(s).

[0074] In some embodiments, offline computation platform 301 can provide an interface that displays parameters organized in a layered structure 400, which can facilitate the parameter value setup process. FIG. 4A is an example parameter structure for parameter setup, consistent with embodiments of the present disclosure. As shown in FIG. 4A. parameter structure 400 can be configured to have multiple levels. In FIG. 4A, each level can include multiple parameter groups belonging to a different category. For example, a first level of parameters LI may include a first group of parameters PG1 related to area selection for an inspection / metrology, a second group of parameters PG2 related to image enhancement, and a third group of parameters PG3 related to defect type. A second level of parameters L2 may include parameter subgroups belonging to each parameter group PG1, PG2, or PG3 in first level LI. In some embodiments, the parameter subgroups belonging to a certain parameter group in first level LI can be configured to show up on the interface after the corresponding parameter group is selected. In FIG. 4A, subgroups of parameters belonging to third group PG3 are shown as third group PG3 is selected, which is indicated as a gray color. For example, second level L2 may include a first subgroup of parameters PSG31 related to a first defect type, asecond subgroup of parameters PSG32 related to a second defect type, and a third subgroup of parameters PSG33 related to a third defect type. Similarly, a third level of parameters L3 may include parameter subgroups belonging to each upper parameter subgroup PSG31, PSG32, or PSG33 in second level L2. In FIG. 4A, subgroups of parameters belonging to second subgroup PSG32 are shown as second subgroup PSG32 is selected, which is indicated as a darker gray color. For example, third level L3 may include a first subgroup of parameters PSG321 related to a first threshold value for determining the second defect type, a second subgroup of parameters PSG322 related to a second threshold value for determining the second defect type, and a third subgroup of parameters PSG323 related to a third threshold value for determining the second defect type.

[0075] While it is illustrated that parameter structure 400 has three levels each having three groups, it will be appreciated that any level with any number of parameter groups / parameters can be utilized in embodiments of the present disclosure. It will also be appreciated that any parameters that can be utilized for inspection / metrology in inline tool 1001 can be set and tested for optimization. In some embodiments, parameter structure 400 may include other parameters that are not utilized in inline tool 1001. For example, when inline tool 1001 uses the POR recipe to detect a first and second defect type, parameter structure 400 may have a parameter to detect a third defect type.

[0076] Referring back to FIG. 3, according to some embodiments of the present disclose, during DOE process 310, an image enhancement (IH) parameter can be setup and applied. During IH process 3110, various image enhancement parameters can be set and applied to input image 3001. FIG. 4B illustrates example enhanced images 3002 after applying various image enhancement parameters to input image 3001, consistent with embodiments of the present disclosure. As shown in FIG. 4B, a first group of images indicate results after applying a first image enhancement parameter El to a group of input images 3001. Similarly, a second group of images indicate results after applying a second image enhancement parameter E2 to the group of input images 3001. FIG. 4B also illustrates an Nthgroup of images after applying an N* image enhancement parameter EN to the group of input images 3001.

[0077] According to some embodiments of the present disclosure, a range of images having different characteristics can be populated by applying various image enhancement parameters to input image 3001. FIG. 4C illustrates example enhanced images 3002 after applying various image enhancement parameters (e.g., El to EN) to an input image 3001, consistent with embodiments of the present disclosure. For example, enhanced image 401 can be obtained by applying a first image enhancement parameter to input image 3001 and enhanced image 402 can be obtained by applying an N* image enhancement parameter to input image 3001. In FIG. 4C, input image 3001 is also included as a raw image that has not undergone any image enhancement. As discussed later with respect to FIG. 5B, FIG. 5C, and FIG. 5D, each enhanced image 3002 shows different characteristics that could make the enhance image more suitable for certain defect inspection or metrology.

[0078] Referring back to FIG. 3, according to some embodiments of the present disclose, during DOE process 310, a detection (DD) parameter can be setup and applied. During DD process 3120, various detection parameters can be set, and defect detection or metrology can be performed on input image 3001 or enhanced image 3002 according to the set detection parameters.

[0079] FIG. 5A illustrates example detection results 3003 after applying various detection parameters to enhanced images, consistent with embodiments of the present disclosure. In FIG. 5A, first column Cl represents results after applying first detection parameter DI to enhanced images, e.g., enhanced images 3002 of FIG. 4B, and second column C2 represents results after applying second detection parameter D2 to enhanced images 3002 of FIG. 4B. Similarly, the Mlhcolumn CM represents results after applying Mlhdetection parameter DM to enhanced images 3002 of FIG. 4B. According to some embodiments of the present disclosure, detection results 3003 in FIG. 5A can be utilized to evaluate performance of each combination of two parameters of an image enhancement parameters (e.g., El to EN) and a detection parameter (e.g., DI to DM).

[0080] According to some embodiments of the present disclosure, each enhanced image 3002 with a different image enhancement parameter from one original input image 3001 can show different characteristics. FIGs. 5B, 5C, and 5D illustrate example enhanced images more suitable for certain defect detection or metrology. FIG. 5B illustrates an example image that is more suitable for critical dimension measurement, consistent with embodiments of the present disclosure. In FIG. 5B, image 510 can be an original image, e.g., input image 3001 without any image enhancement parameter being applied. As shown in FIG. 5B, image 510 may have characteristics that are suitable for measuring a critical dimension 5001 of a structure having a certain pattern. It will be appreciated that image 510 and its critical dimension 5001 measuring are provided as an example for illustration, any enhancement could be applied to input images to make the image suitable for any other types of critical dimensions. FIG. 5C illustrates an example image that is suitable for bridge defect detection, consistent with embodiments of the present disclosure. In FIG. 5C, image 520 can be an enhanced image after applying a certain image enhancement parameter, such as higher contrast, to input image 3001. As shown in FIG. 5C, image 520 may have characteristics that are suitable for detecting a bridge defect 5002. FIG. 5D illustrates an example image that is suitable for broken defect detection, consistent with embodiments of the present disclosure. In FIG. 5D, image 530 can be an enhanced image after applying a certain image enhancement parameter, such as lower brightness, to input image 3001. As shown in FIG. 5D, image 530 may have characteristics that are suitable for detecting a broken defect 5003.

[0081] Referring back to FIG. 3, according to some embodiments of the present disclosure, at DD process 3120, one or more additional detection parameters can be set, and defect detection or metrology can be performed on input image 3001 or enhanced image 3002 according to the set detection parameters. FIG. 6A illustrates example results after applying various threshold parameters for a first defect, consistent with embodiments of the present disclosure. In FIG. 6A, first columnCl 1 represents results after applying first threshold T1 for first detection parameter DI to enhanced images, e.g., enhanced images 3002 of FIG. 4B, and second column C12 represents results after applying second threshold T2 for first detection parameter DI to enhanced images 3002 of FIG. 4B. Similarly, the K* column C1K represents results after applying K* threshold TK for first detection parameter DI to enhanced images 3002 of FIG. 4B. According to some embodiments of the present disclosure, results in FIG. 6A can be utilized to evaluate performance of each combination of three parameters of an image enhancement parameter (e.g., El), first detection parameter (e.g., DI), and various threshold (e.g., T1 to TK).

[0082] According to some embodiments of the present disclosure, a certain defect can be identified when a certain measurement exceeds a set threshold. In some embodiments, a threshold for detecting a certain defect type can be determined based on a most stable critical dimension suitable for a certain patterns. FIG. 6B illustrates effects of a threshold parameter change in detecting a defect, consistent with embodiments of the present disclosure. In FIG. 6B, two bridge defect candidates 6004 and 6005 are illustrated. Depending on an applied threshold, two bridge defect candidates 6004 and 6005 can be identified as a defect or not. For example, a bridge defect can be identified when a distance between two adjacent structures is less than a threshold. When the threshold is set too small, none of two bridge defect candidates 6004 and 6005 can be identified as a bridge defect. In contrast, when the threshold is set too big, both bridge defect candidates 6004 and 6005 can be identified as a bridge defect. When the threshold is set in between, first bridge defect candidate 6004 may not be identified as a bridge defect while second bridge defect candidate 6005 may be identified as a bridge defect. Setting a proper threshold for a certain defect type can improve a defect capture rate and lower a nuisance rate. While embodiments have been illustrated with a limited number of parameter combinations including a limited number of parameters, it will be appreciated any number of parameter combinations including any number of parameters can be utilized in DOE process 310 to generate various experiment results based on which performance of the set parameter combinations are evaluated.

[0083] Referring back to FIG. 3, according to some embodiments of the present disclosure, detection analysis can be performed based on detection results 3003. In some embodiments, detection analysis can be performed for each parameter combination. In some embodiments, during analysis process 320, defects found from detection results 3003 can be listed as defects. In some embodiments, defects found from detection results 3003 can be associated with a corresponding location on a wafer or design data. In some embodiments, during analysis process 320, a nuisance rate can be determined from detection results 3003. For example, the ratio of the number of nuisance defects to the total number of defects found from detection results 3003 can be determined. In some embodiments, the defects that are found from detection results 3003 and that are not nuisance defects can be listed as captured defects. The capture rate can also be determined for each parameter combination during analysis process 320. In some embodiments, captured defects can be grouped according to its defecttype. In some embodiments, a process time for defect detection or metrology can be determined for each parameter combination.

[0084] In some embodiments, defects or their corresponding locations can be identified by comparing detection results 3003 with corresponding design data. In some embodiments, design data can be a layout file for a wafer design, which is a golden image or in a Graphic Database System (GDS) format, Graphic Database System II (GDS II) format, an Open Artwork System Interchange Standard (OASIS) format, a Caltech Intermediate Format (CIF), etc. The wafer design may include patterns or structures for inclusion on the wafer. The patterns or structures can be mask patterns used to transfer features from the photolithography masks or reticles to a wafer. In some embodiments, a layout in GDS or OASIS format, among others, may comprise feature information stored in a binary file format representing planar geometric shapes, text, and other information related to the wafer design. In some embodiments, design data can also be acquired from inline tool 1001 when corresponding input image 3001 is streamed from inline tool 1001.

[0085] In some embodiments, captured defects can be compared to the results from inline tool 1001. In some embodiments, when a new defect is discovered from detection results 3003, which has not been discovered by inline tool 1001, the new defect is indicated as a newly discovered defect. In some embodiments, when a new defect type is discovered from detection results 3003, which has not been discovered by inline tool 1001, the new defect type is indicated as a newly discovered defect type. In some embodiments, when a defect that has been discovered by inline tool 1001 is not discovered from detection results 3003, the defect is indicated as a missing defect.

[0086] FIG. 7A illustrates a first example analysis result for a set parameter combination, consistent with embodiments of the present disclosure. In FIG. 7A, diagram 710 illustrates the detected defects on corresponding locations on a wafer 711. In FIG. 7A, defect 7001 indicates a defect already found by inline tool 1001 and defect 7002 indicates a defect newly discovered by offline computation platform 301 with the set parameter combination. In some embodiments, offline computation platform 301 may not discover existing defect 7001, which has been found by inline tool 1001 and can be referred to as a missing defect in this disclosure. In some embodiments, while not illustrated in FIG. 7A, the missing defect can also be indicated in diagram 710. While diagram 710 illustrates defects of a first defect type, it will be appreciated that similar analysis can be performed with any defect type. FIG. 7B illustrates a second example analysis result for a set parameter combination, consistent with embodiments of the present disclosure. Graph 720 shows the number of new defects 7002 compared to the number of existing defects 7001 for first defect type DI and second defect type D2. In some embodiments, graph 720 can show a ratio of the number of new defects 7002 to the number of existing defects 7001.

[0087] According to some embodiments of the present disclosure, with reference to FIG. 3, analysis results 3004 (such as the results shown in FIG. 7A and FIG. 7B) can be generated for each parameter combination including two or more parameters or for each defect type. In some embodiments,analysis results 3004 can be provided with any other forms that can facilitate qualitative or quantitative evaluation of a parameter combination. According to some embodiments of the present disclosure, analysis results 3004 can also include a nuisance rate for each parameter combination or for each defect type. According to some embodiments of the present disclosure, analysis results 3004 can have a capture rate or a nuisance rate for each defect type or for each parameter combination. According to some embodiments of the present disclosure, analysis results 3004 can include throughput information such as a process time for each defect type or for each parameter combination. According to some embodiments of the present disclosure, analysis process 320 can be performed for input image 3001 generated by a single -beam inspection system (e.g., single-beam tool 104B of FIG. 4B) or for input image 3001 generated by each beam of multi-beam inspection system (e.g., multibeam tool 104A of FIG. 4A). And in some embodiments, analysis results 3004 can be generated for each beam channel (e.g., a single-beam of single-beam tool 104B of FIG. 4B or each beam of multibeam tool 104A of FIG. 4A). According to some embodiments of the present disclosure, a deep learning algorithm can be applied to analysis process 320. In some embodiments, a deep learning algorithm can be utilized to group defects into multiple groups per defect type, to determine a capture rate or a nuisance rate, etc. Because offline computation platform 301, which is separate from inline tool 1001, performs recipe tuning for inline tool 1001, massive data can be analyzed in real time with an aid of computationally intensive deep learning model without increasing wafer cycle time.

[0088] Referring back to FIG. 3, according to some embodiments of the present disclosure, a parameter optimization can be performed based on analysis results 3004. In some embodiments, a parameter combination that is optimized for each defect type detection can be found during optimization process 330. According to some embodiments of the present disclosure, a key performance indicator (KPI) can be setup to evaluate performance of a parameter combination for detecting a certain defect type. For example, the KPI for a certain defect type can indicate a certain range of capture rate, throughput, etc. In some embodiments, a KPI can change depending on a defect type, implementations, requirements, etc. For example, a KPI can be determined on various criteria, e.g., a defect threshold, defect strength, gray level histogram, image sharpness, and so on.

[0089] FIG. 7C illustrates an example KPI map per parameter combination, consistent with embodiments of the present disclosure. In FIG. 7C, the X-axis represents a value for a first parameter Pl, the Y-axis represents a value for a second parameter P2, the graph illustrates a capture rate as a KPI according to a combination of first parameter Pl and second parameter P2. For example, a capture rate at value pl 1 for first parameter Pl and value p21 for second parameter is in capture rate 60-80% in FIG. 7C. Similarly, the graph can indicate a capture rate for any combination of first parameter Pl and second parameter P2.

[0090] According to some embodiments of the present disclosure, parameter combinations can be ranked automatically during optimization process 330 according to its KPI value. For example, parameter combinations for a certain defect type can be ranked according to its capture rate such thata parameter combination having a highest capture rate can be ranked higher. In some embodiments where two parameter combinations have the same capture rate, the parameter combinations can be ranked according to additional criterion. For example, a parameter combination providing a more stable defect detection result can be ranked higher. In FIG. 7C, parameter combinations falling in first area Al and second area A2 have the same capture rate 60-80%. In this example, a first parameter combination of value pl 1 for first parameter Pl and value p21 for second parameter P2 can be ranked higher than a second parameter combination of value pl2 for first parameter Pl and value p22 for second parameter P2 because the first parameter combination is located in a wide area (i.e., Al) than the second parameter combination in second area A2. In this case, when using the first parameter combination, the defect detection result can be more stable even when the parameters are not precisely calibrated. It will be appreciated that various ranking strategies can be applicable to some embodiments of the present disclosure. In some embodiments, parameter combinations that only meet a certain criterion can be automatically ranked. For example, the criterion can be a defect capture rate that is equal to or above 60%. In this example, parameter combinations having capture rate equal to or above 60% can be ranked to be used for inline production.

[0091] According to some embodiments of the present disclosure, a deep learning algorithm can be applied to optimization process 330. In some embodiments, a deep learning algorithm can be trained to determine an optimized parameter combination based on analysis results 3004 and set criteria. In some embodiments, the deep learning algorithm can automatically rank parameter combination according to its set criteria and KPI. Because offline computation platform 301, which is separate from inline tool 1001, performs recipe tuning for inline tool 1001, massive data can be analyzed in real time with an aid of computationally intensive deep learning model without increasing wafer cycle time. According to some embodiments of the present disclosure, optimization results including an optimized parameter combination 3005 or ranked parameter combinations for a certain defect type can be saved in offline computation platform 301, e.g., for future reference. In some embodiments, parameter combinations ranked with different KPIs or criteria (e.g., throughput, etc.) can also be saved in offline computation platform 301. In some embodiments, only certain percentage or certain number of top ranked parameter combinations can be saved in offline computation platform 301.

[0092] According to some embodiments of the present disclosure, a parameter combination 3005 that is determined to be optimized parameter combination for a certain defect type can be provided to inline tool 1001. In some embodiments, a parameter combination 3005 can be utilized in real time inline tool 1001 in detecting defects for the defect type. In some embodiments, inline tool 1001 can update the parameters of the POR recipe corresponding to parameter combination 3005 and continue to perform inspection with the updated parameters. According to some embodiments of the present disclosure, a parameter combination 3005 that is determined to be optimized parameter combination for a certain defect type can also be utilized in DOE process 310. For example, during DOE process 310, parameter ranges for experiments to detect a certain defect type can be reduced based on theprovided parameter combination 3005. Accordingly, more accurate experiment results can be obtained with less experiment time.

[0093] Referring back to FIG. 3, according to some embodiments of the present disclosure, a weak point prediction can be performed based on analysis results 3004. In some embodiments, during prediction process 340, a pattern analysis can be performed to determine pattern(s) that is more likely to have a defect for a certain defect type based on patterns of found defects and design data. The pattern that is likely to have a defect of a certain defect type can be determined based on the analysis results 3004. In some embodiments, a pattern search can be applied to find additional locations on a reticle or a mask pattern that are likely to have a defect of a certain defect type. In some embodiments, a pattern search can be performed to find a location(s) on design data having a pattern that is similar to or same as patterns having a defect of a certain defect type.

[0094] According to some embodiments of the present disclosure, weak point data 3006 can be provided to inline tool 1001. In some embodiments, weak point data 3006 can be utilized by inline tool 1001 to inspect portions of wafers corresponding to weak point locations that are likely to have defects of a certain defect type. In some embodiments, inline tool 1001 can inspect portions of the wafer corresponding to the weak point locations for a certain defect type with parameter combination 3005 that is determined to be optimized parameter combination for the certain defect type.

[0095] According to some embodiments of the present disclosure, because the inspection performance for a certain parameter combination is monitored during real time recipe tuning 300, it is possible to notice HVM process variations or inspection tool drifting in early stage for fixation. Further, according to some embodiments of the present disclosure, because tuned parameter combinations are all saved, it is also possible to provide tool-to-tool matching for better accuracy and consistency in defect inspection or metrology. For example, the POR recipe of an inspection tool can be compared and verified between different runs on the same tool by comparing performance of the different runs. In some embodiments where multi-beams are used, the POR recipe of each beam can be compared and verified between two different beams. In some embodiments where two different systems are used, the POR recipe can be compared and verified between two different tools.

[0096] FIG. 8 is a flow chart illustrating an example method for real time recipe tuning of an inline inspection system, consistent with embodiments of the present disclosure. The steps of method 800 can be performed by an offline computation platform, e.g., offline computation platform 301 of FIG. 3. It is appreciated that the illustrated method 800 can be altered to modify the order of steps and to include additional steps.

[0097] In step S810, one or more input images can be acquired from a separate wafer inspection system. In some embodiments, wafer inspection system can be inline inspection tool, e.g., inline tool 1001 of FIG. 3. In some embodiments, input images can be input image 3001 of FIG. 3. In some embodiments, input image 3001 is streamed from inline tool 1001 in real time as it is taken from inline tool 1001. According to some embodiments of the present disclosure, when acquiring inputimage 3001, the POR recipe including parameter information used for generating input image 3001 can also be acquired from inline tool 1001. In some embodiments, inline tool’s image inspection and metrology results for input image 3001 can also be acquired along with the POR parameter recipe for performing the image inspection and metrology on input image 3001.

[0098] In step S820, a parameter recipe can be set up and detection results according to the set parameter recipe can be generated. In some embodiments, experiment results can be detection results 3003 in FIG. 3. In step S820, parameter values for a set of parameters of interest can be setup for experiments. In some embodiments, a set of parameters can include at least one parameter that is utilized in inline tool 1001 for inspection and metrology of an inspection image. In some embodiments, step S820 can be iterated to sweep through various combinations of parameters of interest. In some embodiments, experiments with pre-designed or pre-determined recipe combination(s) can be automatically triggered as input image 3001 is streamed to offline computation platform 301. According to some embodiments of the present disclosure, input image 3001 is used as input for image inspection for defect detection or metrology with the pre-determined recipe combination(s).

[0099] In some embodiments, in step S820, an image enhancement (IH) parameter can be setup and applied. In step S820, various image enhancement parameters can be set and applied to input image 3001. According to some embodiments of the present disclosure, a range of images having different characteristics can be populated by applying various image enhancement parameters to input image 3001. In some embodiments, in step S820, a detection parameter can be setup and applied. In step S820, various detection parameters can be set, and defect detection or metrology can be performed on input image 3001 or enhanced image 3002 according to the set detection parameters. In some embodiments, detection parameters can include a dimension measurement criterion parameter, a defect type parameter, a threshold parameter for detecting a certain defect type, etc. In step S820, detection results can include critical dimension measurements or defects found on the images according to the set parameter recipe.

[0100] In step S830, detection analysis can be performed based on detection results. In some embodiments, detection analysis can be performed for each parameter combination. In some embodiments, in step S830, defects found from detection results 3003 can be listed as defects. In some embodiments, defects found from detection results 3003 can be associated with a corresponding location on a wafer or design data. In some embodiments, in step S830, a nuisance rate can be determined from detection results 3003. In some embodiments, the defects that are found from detection results 3003 and that are not nuisance defects can be listed as captured defects. The capture rate can also be determined for each parameter combination during analysis process 320. In some embodiments, captured defects can be grouped according to its defect type. In some embodiments, a process time for defect detection or metrology can be determined for each parameter combination. In some embodiments, captured defects can be compared to the results from inline tool 1001. In someembodiments, when a new defect is discovered from detection results 3003, which has not been discovered by inline tool 1001, the new defect is indicated as a newly discovered defect. In some embodiments, when a new defect type is discovered from detection results 3003, which has not been discovered by inline tool 1001, the new defect type is indicated as a newly discovered defect type. In some embodiments, when a defect that has been discovered by inline tool 1001 is not discovered from detection results 3003, the defect is indicated as a missing defect.

[0101] In step S840, a parameter optimization can be performed based on analysis results from step S830. In some embodiments, analysis results can be analysis results 3004 in FIG. 3. In some embodiments, a parameter combination that is optimized for each defect type detection can be found in step S840. According to some embodiments of the present disclosure, a key performance indicator (KPI) can be setup to evaluate performance of a parameter combination for detecting a certain defect type. For example, the KPI for a certain defect type can indicate a certain range of capture rate, throughput, etc. In some embodiments, a KPI can change depending on a defect type, implementations, requirements, etc. In some embodiments, in step S840, parameter combinations can be ranked automatically according to its KPI value. For example, parameter combinations for a certain defect type can be ranked according to its capture rate such that a parameter combination having a highest capture rate can be ranked higher. It will be appreciated that various ranking strategies can be applicable to some embodiments of the present disclosure. In some embodiments, parameter combinations that only meet a certain criterion can be automatically ranked.

[0102] In step S860, a parameter combination that is determined to be optimized parameter combination for a certain defect type can be provided to inline tool 1001. In some embodiments, the parameter combination can be a parameter combination 3005 of FIG. 3. In some embodiments, parameter combination 3005 can be utilized in real time inline tool 1001 in detecting defects for the defect type. In some embodiments, inline tool 1001 can update the parameters of the POR recipe corresponding to parameter combination 3005 and continue to perform inspection with the updated parameters. According to some embodiments of the present disclosure, a parameter combination 3005 that is determined to be optimized parameter combination for a certain defect type can also be utilized in step S820. For example, in step S820, parameter ranges for experiments to detect a certain defect type can be reduced based on the provided parameter combination 3005.

[0103] In step S850, a weak point prediction can be performed. In some embodiments, in step S850, a pattern analysis can be performed to determine pattern(s) that is more likely to have a defect for a certain defect type based on patterns of found defects and design data. The pattern that is likely to have a defect of a certain defect type can be determined based on the analysis results 3004. In some embodiments, a pattern search can be applied to find additional locations on a reticle or a mask pattern that are likely to have a defect of a certain defect type. In some embodiments, a pattern search can be performed to find a location(s) on design data having a pattern that is similar to or same as patterns having a defect of a certain defect type.

[0104] In step S860, weak point data can be provided to inline tool 1001. In some embodiments, weak point data can be utilized by inline tool 1001 to inspect portions of wafers corresponding to weak point locations that are likely to have defects of a certain defect type. In some embodiments, inline tool 1001 can inspect portions of the wafer corresponding to the weak point locations for a certain defect type with parameter combination 3005 that is determined to be optimized parameter combination for the certain defect type.

[0105] FIG. 9 is a block diagram that illustrates an example computer system 900 that can assist in implementing the methods, flows or the apparatus disclosed herein. In some embodiments, computer system 900 can assist in implementing at least part of controller 109 of FIG. 1, image processing system 290 of FIG. 2A, or image processing system 199 of FIG. 2B. In some embodiments, computer system 900 can assist in implementing at least part of processes or offline computation platform 301 of FIG. 3 or method 800 of FIG. 8. Computer system 900 includes a bus 902 or other communication mechanism for communicating information, and a processor 904 (or multiple processors 904 and 905, which can include hardware accelerators) coupled with bus 902 for processing information. Computer system 900 also includes a main memory 906, such as a random access memory (RAM) or other dynamic storage device, coupled to bus 902 for storing information and instructions to be executed by processor 904. Main memory 906 also may be used for storing temporary variables or other intermediate information during execution of instructions to be executed by processor 904. Computer system 900 further includes a read only memory (ROM) 908 or other static storage device coupled to bus 902 for storing static information and instructions for processor 904. A storage device 910, such as a magnetic disk or optical disk, is provided and coupled to bus 902 for storing information and instructions.

[0106] Computer system 900 may be coupled via bus 902 to a display 912, such as a cathode ray tube (CRT) or flat panel or touch panel display for displaying information to a computer user. An input device 914, including alphanumeric and other keys, is coupled to bus 902 for communicating information and command selections to processor 904. Another type of user input device is cursor control 916, such as a mouse, a trackball, or cursor direction keys for communicating direction information and command selections to processor 904 and for controlling cursor movement on display 912. This input device typically has two degrees of freedom in two axes, a first axis (e.g., x) and a second axis (e.g., y), that allows the device to specify positions in a plane. A touch panel (screen) display may also be used as an input device.

[0107] According to some embodiments, portions of one or more methods described herein may be performed by computer system 900 in response to processor 904 executing one or more sequences of one or more instructions contained in main memory 906. Such instructions may be read into main memory 906 from another computer-readable medium, such as storage device 910. Execution of the sequences of instructions contained in main memory 906 causes processor 904 to perform the process steps described herein. One or more processors in a multi-processing arrangement may also beemployed to execute the sequences of instructions contained in main memory 906. In an alternative embodiment, hard-wired circuitry may be used in place of or in combination with software instructions. Thus, the description herein is not limited to any specific combination of hardware circuitry and software.

[0108] The term “computer-readable medium” as used herein refers to any medium that participates in providing instructions to processor 904 for execution. The computer-readable medium includes a non-transitory computer medium. The computer readable medium may store instructions for a processor of a controller (e.g., controller 109 of FIG. 1), a processor of an image processing system 290 of FIG. 2A, a processor of image processing system 199 of FIG. 2B, or a processor of offline computation platform 301 of FIG. 3 to carry out, among other things, image inspection, image acquisition, stage positioning, beam focusing, electric field adjustment, beam bending, condenser lens adjusting, activating charged-particle source, beam deflecting, method 300 of FIG. 3, and method 800 in FIG. 8. Such a medium may take many forms, including but not limited to, non-volatile media, volatile media, and transmission media. Non-volatile media include, for example, optical or magnetic disks, such as storage device 910. Volatile media include dynamic memory, such as main memory 906. Transmission media include coaxial cables, copper wire and fiber optics, including the wires that comprise bus 902. Transmission media can also take the form of acoustic or light waves, such as those generated during radio frequency (RF) and infrared (IR) data communications.Common forms of computer-readable media include, for example, a floppy disk, a flexible disk, solid state drive, hard disk, magnetic tape, any other magnetic data storage medium, a Compact Disc Read Only Memory (CD-ROM,) DVD, any other optical data storage medium, punch cards, paper tape, any other physical medium with patterns of holes, a Random Access Memory (RAM), a Programmable Read Only Memory (PROM), and Erasable Programmable Read Only Memory (EPROM), a FLASH- EPROM, any other flash memory, Non-Volatile Random Access Memory (NVRAM), a cache, a register, any other memory chip or cartridge, a carrier wave as described hereinafter, or any other medium from which a computer can read, and networked versions of the same.

[0109] Various forms of computer readable media may be involved in carrying one or more sequences of one or more instructions to processor 904 for execution. For example, the instructions may initially be borne on a magnetic disk of a remote computer. The remote computer can load the instructions into its dynamic memory and send the instructions over a telephone line using a modem. A modem local to computer system 900 can receive the data on the telephone line and use an infrared transmitter to convert the data to an infrared signal. An infrared detector coupled to bus 902 can receive the data carried in the infrared signal and place the data on bus 902. Bus 902 carries the data to main memory 906, from which processor 904 retrieves and executes the instructions. The instructions received by main memory 906 may optionally be stored on storage device 910 either before or after execution by processor 904.

[0110] Computer system 900 may also include a communication interface 918 coupled to bus 902. Communication interface 918 provides a two-way data communication coupling to a network link 920 that is connected to a local network 922. For example, communication interface 918 may be an integrated services digital network (ISDN) card or a modem to provide a data communication connection to a corresponding type of telephone line. As another example, communication interface 918 may be a local area network (LAN) card to provide a data communication connection to a compatible LAN. Wireless links may also be implemented. In any such implementation, communication interface 918 sends and receives electrical, electromagnetic or optical signals that carry digital data streams representing various types of information.

[0111] Network link 920 typically provides data communication through one or more networks to other data devices. For example, network link 920 may provide a connection through local network 922 to a host computer 924 or to data equipment operated by an Internet Service Provider (ISP) 926. ISP 926 in turn provides data communication services through the worldwide packet data communication network, now commonly referred to as the “Internet” 928. Local network 922 and Internet 928 both use electrical, electromagnetic or optical signals that carry digital data streams. The signals through the various networks and the signals on network link 920 and through communication interface 918, which carry the digital data to and from computer system 900, are exemplary forms of carrier waves transporting the information.

[0112] Computer system 900 can send messages and receive data, including program code, through the network(s), network link 920, and communication interface 918. In the Internet example, a server 930 might transmit a requested code for an application program through Internet 928, ISP 926, local network 922 and communication interface 918. One such downloaded application may provide all or part of a method described herein, for example. The received code may be executed by processor 904 as it is received, or stored in storage device 910, or other non-volatile storage for later execution. In this manner, computer system 900 may obtain application code in the form of a carrier wave.

[0113] The embodiments may further be described using the following clauses:1. A method for real time parameter tuning for a wafer inspection system, comprising: acquiring one or more input images streamed from the wafer inspection system; applying a plurality of image enhancement parameters to the one or more input images to generate a plurality of images with different characteristics; identifying defects from the plurality of images by applying a plurality of defect detection parameters; and determining, based on the defects identified on the plurality of images, a parameter combination of the plurality of image enhancement parameters and of the plurality of the defect detection parameters to be used for detecting a first defect type.2. The method of clause 1 , further comprising:identifying a newly discovered defect that has not been previously detected by the wafer inspection system; and identifying a missing defect that has been detected by the wafer inspection system and that has not been previously identified from the plurality of images.3. The method of clause 1 or 2, wherein identifying the defects from the plurality of images comprises: identifying the defects and corresponding locations of the defects.4. The method of any one of clauses 1-3, further comprising: grouping the defects by defect type.5. The method of clause 4, wherein determining the parameter combination comprises: determining the parameter combination based on the grouping of the defects by defect type.6. The method of any one of clauses 1-5, wherein determining the parameter combination comprises: determining the parameter combination using a deep learning algorithm.7. The method of any one of clauses 1-6, further comprising: providing the determined parameter combination to the wafer inspection system such that the wafer inspection system updates a process of recode recipe in real time according to the determined parameter combination for inline production.8. The method of any one of clauses 1-7, further comprising: predicting a weak point on design data based on the defects identified on the plurality of images.9. The method of clause 8, wherein predicting the weak point on design data comprises: predicting the weak point on the design data by applying a pattern search based on pattern data of the defects identified on the plurality of images.10. The method of clause 8 or 9, further comprising: providing information of the weak point to the wafer inspection system such that the wafer inspection system updates a process of recode recipe in real time to inspect a portion of the wafer corresponding to the weak point.11. The method of any one of clauses 1-10, further comprising: setting a key performance indicator (KPI); and ranking various parameter combinations of the plurality of image enhancement parameters and the plurality of the defect detection parameters according to corresponding KPI values that are determined based on the defects identified on the plurality of images.12. The method of clause 11, wherein the KPI is at least a defect capture rate or a throughput.13. The method of any one of clauses 1-12, wherein the method is performed in parallel while the wafer inspection system is in production.14. The method of any one of clauses 1-13, wherein the wafer inspection system is a separate wafer inspection system.15. A method for real time parameter tuning for a wafer inspection system, comprising: acquiring one or more input images streamed from the wafer inspection system; applying a first parameter combination to the one or more input images and identifying a first set of defects of the one or more input images; applying a second parameter combination to the one or more input images and identifying a second set of defects of the one or more input images; determining a first key performance indicator (KPI) value for the first parameter combination based on the first set of defects; determining a second KPI value for the second parameter combination based on the second set of defects; and ranking the first parameter combination and the second parameter combination based on the first KPI value and the second KPI value.16. The method of clause 15, further comprising: identifying a newly discovered defect that has not been previously detected by the wafer inspection system among the first set of defects; and identifying a missing defect that has been detected by the wafer inspection system and that has not been previously identified by the first set of defects.17. The method of clause 15 or 16, wherein identifying the first set of defects comprises: identifying the first set of defects and their corresponding locations.18. The method of any one of clauses 15-17, further comprising: grouping the first set of defects by defect type.19. The method of clause 18, wherein determining the parameter combination comprises: determining the parameter combination based on the grouping of the defects by defect type.20. The method of clause 19, wherein the first KPI value and the second KPI value are associated with a first defect type.21. The method of any one of clauses 15-20, wherein ranking the first parameter combination and the second parameter combination comprises: ranking the first parameter combination and the second parameter combination using a deep learning algorithm.22. The method of any one of clauses 15-21, further comprising: providing a higher ranked parameter combination between the first parameter combination and the second parameter combination to the wafer inspection system such that the wafer inspection system updates a process of recode recipe in real time according to the provided parameter combination for inline production.23. The method of any one of clauses 15-22, further comprising: predicting a weak point on design data based on the first set of defects.24. The method of clause 23, wherein predicting the weak point on design data comprises: predicting the weak point on the design data by applying a pattern search based on pattern data of the first set of defects.25. The method of clause 23 or 24, further comprising: providing information of the weak point to the wafer inspection system such that the wafer inspection system updates a process of recode recipe in real time to inspect a portion of the wafer corresponding to the weak point.26. The method of any one of clauses 15-25, wherein the KPI is at least a defect capture rate or a throughput.27. The method of any one of clauses 15-26, wherein the wafer inspection system is a separate wafer inspection system.28. An apparatus for real time parameter tuning for a wafer inspection system, comprising: a memory storing a set of instructions; and at least one processor configured to execute the set of instructions to cause the apparatus to perform operations comprising: acquiring one or more input images streamed from the wafer inspection system; applying a plurality of image enhancement parameters to the one or more input images to generate a plurality of images with different characteristics; identifying defects from the plurality of images by applying a plurality of defect detection parameters; and determining, based on the defects identified on the plurality of images, a parameter combination of the plurality of image enhancement parameters and of the plurality of the defect detection parameters to be used for detecting a first defect type.29. The apparatus of clause 28, wherein the operations further comprise: identifying a newly discovered defect that has not been previously detected by the wafer inspection system; and identifying a missing defect that has been detected by the wafer inspection system and that has not been previously identified from the plurality of images.30. The apparatus of clause 28 or 29, wherein in identifying the defects from the plurality of images, the at least one processor is configured to execute the set of instructions to cause the apparatus to further perform: identifying the defects and corresponding locations of the defects.31. The apparatus of any one of clauses 28-30, wherein the operations further comprise: grouping the defects by defect type.32. The apparatus of clause 31, wherein in determining the parameter combination, the at least one processor is configured to execute the set of instructions to cause the apparatus to further perform: determining the parameter combination based on the grouping of the defects by defect type.33. The apparatus of any one of clauses 28-32, wherein in determining the parameter combination, the at least one processor is configured to execute the set of instructions to cause the apparatus to further perform: determining the parameter combination using a deep learning algorithm.34. The apparatus of any one of clauses 28-33, wherein the operations further comprise: providing the determined parameter combination to the wafer inspection system such that the wafer inspection system updates a process of recode recipe in real time according to the determined parameter combination for inline production.35. The apparatus of any one of clauses 28-34, wherein the operations further comprise: predicting a weak point on design data based on the defects identified on the plurality of images.36. The apparatus of clause 35, wherein in predicting the weak point on design data, the at least one processor is configured to execute the set of instructions to cause the apparatus to further perform: predicting the weak point on the design data by applying a pattern search based on pattern data of the defects identified on the plurality of images.37. The apparatus of clause 35 or 36, wherein the operations further comprise: providing information of the weak point to the wafer inspection system such that the wafer inspection system updates a process of recode recipe in real time to inspect a portion of the wafer corresponding to the weak point.38. The apparatus of any one of clauses 28-37, wherein the operations further comprise: setting a key performance indicator (KPI); and ranking various parameter combinations of the plurality of image enhancement parameters and the plurality of the defect detection parameters according to corresponding KPI values that are determined based on the defects identified on the plurality of images.39. The apparatus of clause 38, wherein the KPI is at least a defect capture rate or a throughput.40. The apparatus of any one of clauses 28-39, wherein the at least one processor is configured to execute the set of instructions to cause the apparatus to perform the operations in parallel while the wafer inspection system is in production.41. The apparatus of any one of clauses 28-40, wherein the wafer inspection system is separate from the apparatus.42. An apparatus for real time parameter tuning for a wafer inspection system, comprising: a memory storing a set of instructions; and at least one processor configured to execute the set of instructions to cause the apparatus to perform operations comprising:acquiring one or more input images streamed from the wafer inspection system; applying a first parameter combination to the one or more input images and identifying a first set of defects of the one or more input images; applying a second parameter combination to the one or more input images and identifying a second set of defects of the one or more input images; determining a first key performance indicator (KPI) value for the first parameter combination based on the first set of defects; determining a second KPI value for the second parameter combination based on the second set of defects; and ranking the first parameter combination and the second parameter combination based on the first KPI value and the second KPI value.43. The apparatus of clause 42, wherein the operations further comprise: identifying a newly discovered defect that has not been previously detected by the wafer inspection system among the first set of defects; and identifying a missing defect that has been detected by the wafer inspection system and that has not been previously identified by the first set of defects.44. The apparatus of clause 42 or 43, wherein in identifying the first set of defects, the at least one processor is configured to execute the set of instructions to cause the apparatus to further perform: identifying the first set of defects and their corresponding locations.45. The apparatus of any one of clauses 42-44, wherein the operations further comprise: grouping the first set of defects by defect type.46. The apparatus of clause 45, wherein in determining the parameter combination, the at least one processor is configured to execute the set of instructions to cause the apparatus to further perform: determining the parameter combination based on the grouping of the defects by defect type.47. The apparatus of clause 46, wherein the first KPI value and the second KPI value are associated with a first defect type.48. The apparatus of any one of clauses 42-47, wherein in ranking the first parameter combination and the second parameter combination, the at least one processor is configured to execute the set of instructions to cause the apparatus to further perform: ranking the first parameter combination and the second parameter combination using a deep learning algorithm.49. The apparatus of any one of clauses 42-48, wherein the operations further comprise: providing a higher ranked parameter combination between the first parameter combination and the second parameter combination to the wafer inspection system such that the wafer inspection system updates a process of recode recipe in real time according to the provided parameter combination for inline production.50. The apparatus of any one of clauses 42-49, wherein the operations further comprise:predicting a weak point on design data based on the first set of defects.51. The apparatus of clause 50, wherein in predicting the weak point on design data, the at least one processor is configured to execute the set of instructions to cause the apparatus to further perform: predicting the weak point on the design data by applying a pattern search based on pattern data of the first set of defects.52. The apparatus of clause 50 or 51, wherein the operations further comprise: providing information of the weak point to the wafer inspection system such that the wafer inspection system updates a process of recode recipe in real time to inspect a portion of the wafer corresponding to the weak point.53. The apparatus of any one of clauses 42-52, wherein the KPI is at least a defect capture rate or a throughput.54. The apparatus of any one of clauses 42-53, wherein the wafer inspection system is separate from the apparatus.55. A non-transitory computer readable medium that stores a set of instructions that is executable by at least on processor of a computing device to cause the computing device to perform a method for real time parameter tuning for a wafer inspection system, the method comprising: acquiring one or more input images streamed from the wafer inspection system; applying a plurality of image enhancement parameters to the one or more input images to generate a plurality of images with different characteristics; identifying defects from the plurality of images by applying a plurality of defect detection parameters; and determining, based on the defects identified on the plurality of images, a parameter combination of the plurality of image enhancement parameters and of the plurality of the defect detection parameters to be used for detecting a first defect type.56. The computer readable medium of clause 55, wherein the set of instructions that is executable by at least one processor of the computing device cause the computing device to further perform: identifying a newly discovered defect that has not been previously detected by the wafer inspection system; and identifying a missing defect that has been detected by the wafer inspection system and that has not been previously identified from the plurality of images.57. The computer readable medium of clause 55 or 56, wherein in identifying the defects from the plurality of images, the set of instructions that is executable by at least one processor of the computing device cause the computing device to perform: identifying the defects and corresponding locations of the defects.58. The computer readable medium of any one of clauses 55-57, wherein the set of instructions that is executable by at least one processor of the computing device cause the computing device to further perform:grouping the defects by defect type.59. The computer readable medium of clause 58, wherein in determining the parameter combination, the set of instructions that is executable by at least one processor of the computing device cause the computing device to perform: determining the parameter combination based on the grouping of the defects by defect type.60. The computer readable medium of any one of clauses 55-59, wherein in determining the parameter combination, the set of instructions that is executable by at least one processor of the computing device cause the computing device to perform: determining the parameter combination using a deep learning algorithm.61. The computer readable medium of any one of clauses 55-60, wherein the set of instructions that is executable by at least one processor of the computing device cause the computing device to further perform: providing the determined parameter combination to the wafer inspection system such that the wafer inspection system updates a process of recode recipe in real time according to the determined parameter combination for inline production.62. The computer readable medium of any one of clauses 55-61, wherein the set of instructions that is executable by at least one processor of the computing device cause the computing device to further perform: predicting a weak point on design data based on the defects identified on the plurality of images.63. The computer readable medium of clause 62, wherein in predicting the weak point on design data, the set of instructions that is executable by at least one processor of the computing device cause the computing device to further perform: predicting the weak point on the design data by applying a pattern search based on pattern data of the defects identified on the plurality of images.64. The computer readable medium of clauses 62 or 63, wherein the set of instructions that is executable by at least one processor of the computing device cause the computing device to further perform: providing information of the weak point to the wafer inspection system such that the wafer inspection system updates a process of recode recipe in real time to inspect a portion of the wafer corresponding to the weak point.65. The computer readable medium of any one of clauses 55-64, wherein the set of instructions that is executable by at least one processor of the computing device cause the computing device to further perform: setting a key performance indicator (KPI); andranking various parameter combinations of the plurality of image enhancement parameters and the plurality of the defect detection parameters according to corresponding KPI values that are determined based on the defects identified on the plurality of images.66. The computer readable medium of clause 65, wherein the KPI is at least a defect capture rate or a throughput.67. The computer readable medium of any one of clauses 55-66, wherein the set of instructions that is executable by at least one processor of the computing device cause the computing device to perform the method in parallel while the wafer inspection system is in production.68. The computer readable medium of any one of clauses 55-67, wherein the wafer inspection system is a separate wafer inspection system.69. A non-transitory computer readable medium that stores a set of instructions that is executable by at least on processor of a computing device to cause the computing device to perform a method for real time parameter tuning for a wafer inspection system, the method comprises: acquiring one or more input images streamed from the wafer inspection system; applying a first parameter combination to the one or more input images and identifying a first set of defects of the one or more input images; applying a second parameter combination to the one or more input images and identifying a second set of defects of the one or more input images; determining a first key performance indicator (KPI) value for the first parameter combination based on the first set of defects; determining a second KPI value for the second parameter combination based on the second set of defects; and ranking the first parameter combination and the second parameter combination based on the first KPI value and the second KPI value.70. The computer readable medium of clause 69, wherein the set of instructions that is executable by at least one processor of the computing device cause the computing device to further perform: identifying a newly discovered defect that has not been previously detected by the wafer inspection system among the first set of defects; and identifying a missing defect that has been detected by the wafer inspection system and that has not been previously identified by the first set of defects.71. The computer readable medium of clause 69 or 70, wherein in identifying the first set of defects, the set of instructions that is executable by at least one processor of the computing device cause the computing device to perform: identifying the first set of defects and their corresponding locations.72. The computer readable medium of any one of clauses 69-71, wherein the set of instructions that is executable by at least one processor of the computing device cause the computing device to further perform:grouping the first set of defects by defect type.73. The computer readable medium of clause 72, wherein in determining the parameter combination, the set of instructions that is executable by at least one processor of the computing device cause the computing device to perform: determining the parameter combination based on the grouping of the defects by defect type.74. The computer readable medium of clause 73, wherein the first KPI value and the second KPI value are associated with a first defect type.75. The computer readable medium of any one of clauses 69-74, wherein in ranking the first parameter combination and the second parameter combination, the set of instructions that is executable by at least one processor of the computing device cause the computing device to perform: ranking the first parameter combination and the second parameter combination using a deep learning algorithm.76. The computer readable medium of any one of clauses 69-75, wherein the set of instructions that is executable by at least one processor of the computing device cause the computing device to further perform: providing a higher ranked parameter combination between the first parameter combination and the second parameter combination to the wafer inspection system such that the wafer inspection system updates a process of recode recipe in real time according to the provided parameter combination for inline production.77. The computer readable medium of any one of clauses 69-76, wherein the set of instructions that is executable by at least one processor of the computing device cause the computing device to further perform: predicting a weak point on design data based on the first set of defects.78. The computer readable medium of clause 77, wherein in predicting the weak point on design data, the set of instructions that is executable by at least one processor of the computing device cause the computing device to perform: predicting the weak point on the design data by applying a pattern search based on pattern data of the first set of defects.79. The computer readable medium of clause 78 or 79, wherein the set of instructions that is executable by at least one processor of the computing device cause the computing device to further perform: providing information of the weak point to the wafer inspection system such that the wafer inspection system updates a process of recode recipe in real time to inspect a portion of the wafer corresponding to the weak point.80. The computer readable medium of any one of clauses 69-79, wherein the KPI is at least a defect capture rate or a throughput.81. The computer readable medium of any one of clauses 69-80, wherein the wafer inspection system is a separate wafer inspection system.

[0114] Block diagrams in the figures may illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer hardware or software products according to various exemplary embodiments of the present disclosure. In this regard, each block in a schematic diagram may represent certain arithmetical or logical operation processing that may be implemented using hardware such as an electronic circuit. Blocks may also represent a module, segment, or portion of code that comprises one or more executable instructions for implementing the specified logical functions. It should be understood that in some alternative implementations, functions indicated in a block may occur out of the order noted in the figures. For example, two blocks shown in succession may be executed or implemented substantially concurrently, or two blocks may sometimes be executed in reverse order, depending upon the functionality involved.Some blocks may also be omitted. It should also be understood that each block of the block diagrams, and combination of the blocks, may be implemented by special purpose hardware-based systems that perform the specified functions or acts, or by combinations of special purpose hardware and computer instructions.

[0115] It will be appreciated that the embodiments of the present disclosure are not limited to the exact construction that has been described above and illustrated in the accompanying drawings, and that various modifications and changes may be made without departing from the scope thereof. The present disclosure has been described in connection with various embodiments, other embodiments of the invention will be apparent to those skilled in the art from consideration of the specification and practice of the invention disclosed herein. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the invention being indicated by the following claims.

Claims

CLAIMS1. A method for real time parameter tuning for a wafer inspection system, comprising: acquiring one or more input images streamed from the wafer inspection system; applying a plurality of image enhancement parameters to the one or more input images to generate a plurality of images with different characteristics; identifying defects from the plurality of images by applying a plurality of defect detection parameters; and determining, based on the defects identified on the plurality of images, a parameter combination of the plurality of image enhancement parameters and of the plurality of the defect detection parameters to be used for detecting a first defect type.

2. The method of claim 1, further comprising: identifying a newly discovered defect that has not been previously detected by the wafer inspection system; and identifying a missing defect that has been detected by the wafer inspection system and that has not been previously identified from the plurality of images.

3. The method of claim 1, wherein determining the parameter combination comprises: determining the parameter combination using a deep learning algorithm.

4. The method of claim 1, further comprising: providing the determined parameter combination to the wafer inspection system such that the wafer inspection system updates a process of recode recipe in real time according to the determined parameter combination for inline production.

5. The method of claim 1, further comprising: predicting a weak point on design data based on the defects identified on the plurality of images.

6. The method of claim 5, wherein predicting the weak point on design data comprises: predicting the weak point on the design data by applying a pattern search based on pattern data of the defects identified on the plurality of images.

7. The method of claim 5, further comprising:providing information of the weak point to the wafer inspection system such that the wafer inspection system updates a process of recode recipe in real time to inspect a portion of the wafer corresponding to the weak point.

8. The method of claim 1, further comprising: setting a key performance indicator (KPI); and ranking various parameter combinations of the plurality of image enhancement parameters and the plurality of the defect detection parameters according to corresponding KPI values that are determined based on the defects identified on the plurality of images.

9. A method for real time parameter tuning for a wafer inspection system, comprising: acquiring one or more input images streamed from the wafer inspection system; applying a first parameter combination to the one or more input images and identifying a first set of defects of the one or more input images; applying a second parameter combination to the one or more input images and identifying a second set of defects of the one or more input images; determining a first key performance indicator (KPI) value for the first parameter combination based on the first set of defects; determining a second KPI value for the second parameter combination based on the second set of defects; and ranking the first parameter combination and the second parameter combination based on the first KPI value and the second KPI value.

10. The method of claim 9, further comprising: identifying a newly discovered defect that has not been previously detected by the wafer inspection system among the first set of defects; and identifying a missing defect that has been detected by the wafer inspection system and that has not been previously identified by the first set of defects.

11. The method of claim 9, wherein ranking the first parameter combination and the second parameter combination comprises: ranking the first parameter combination and the second parameter combination using a deep learning algorithm.

12. The method of claim 9, further comprising: providing a higher ranked parameter combination between the first parameter combination and the second parameter combination to the wafer inspection system such that the wafer inspectionsystem updates a process of recode recipe in real time according to the provided parameter combination for inline production.

13. The method of claim 9, further comprising: predicting a weak point on design data based on the first set of defects.

14. The method of claim 13, further comprising: providing information of the weak point to the wafer inspection system such that the wafer inspection system updates a process of recode recipe in real time to inspect a portion of the wafer corresponding to the weak point.

15. An apparatus for real time parameter tuning for a wafer inspection system, comprising: a memory storing a set of instructions; and at least one processor configured to execute the set of instructions to cause the apparatus to perform operations comprising: acquiring one or more input images streamed from the wafer inspection system; applying a plurality of image enhancement parameters to the one or more input images to generate a plurality of images with different characteristics; identifying defects from the plurality of images by applying a plurality of defect detection parameters; and determining, based on the defects identified on the plurality of images, a parameter combination of the plurality of image enhancement parameters and of the plurality of the defect detection parameters to be used for detecting a first defect type.

Images