WO2020077784A1 - 判断缺陷叠图聚集的方法及其系统 - Google Patents

判断缺陷叠图聚集的方法及其系统 Download PDF

Info

Publication number
WO2020077784A1
WO2020077784A1 PCT/CN2018/120979 CN2018120979W WO2020077784A1 WO 2020077784 A1 WO2020077784 A1 WO 2020077784A1 CN 2018120979 W CN2018120979 W CN 2018120979W WO 2020077784 A1 WO2020077784 A1 WO 2020077784A1
Authority
WO
WIPO (PCT)
Prior art keywords
defect
overlay
defects
batch
products
Prior art date
Application number
PCT/CN2018/120979
Other languages
English (en)
French (fr)
Inventor
高嵩
Original Assignee
深圳市华星光电半导体显示技术有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 深圳市华星光电半导体显示技术有限公司 filed Critical 深圳市华星光电半导体显示技术有限公司
Publication of WO2020077784A1 publication Critical patent/WO2020077784A1/zh

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/23—Clustering techniques

Definitions

  • the present application relates to the technical field of display manufacturing, and in particular to a method and system for judging defect overlays.
  • the purpose of the present application is to provide a method for judging the defect overlapping image aggregation, so as to realize the automatic judgment of the defect overlapping image aggregation, so as to quickly find the defect overlapping image aggregation phenomenon.
  • a method for judging the aggregation of defect overlays includes the following steps:
  • the defect overlay is obtained by overlaying defects of a batch of products with the same defect, and there are at least two of the same defects in the defect group;
  • defect overlay is aggregated
  • the screening out the defect group in the defect overlay includes the following steps:
  • the any two defects in the defect stack is less than the second preset value, the any two defects belong to the same defect group.
  • the second preset value is 280-320 microns.
  • the second preset value is 300 ⁇ m.
  • the method for determining the defect overlay image aggregation further includes tracing the source of the defect overlay image aggregation, and tracing the source of the defect overlay image aggregation includes the following steps:
  • the identification code records the For defect information, the processing equipment includes a single-path machine and a reaction chamber.
  • the processing equipment of the batch of products corresponding to the aggregation of defective overlays is a single-path machine
  • the single-path machine is extracted from the defective overlay
  • Defects of the batch of products are stacked together, and the alarm message that the single path machine causes the defects to be stacked is issued at the same time;
  • the processing equipment of the batch of products corresponding to the stacking of defective stacks is a reaction chamber
  • the defects of the batch of products produced in the reaction chamber are extracted from the stack of defectives and stacked, and the Describe the alarm message in the reaction chamber that causes defects to stack up.
  • the defect information includes defect information of the first type and defect information of the second type, and the defect overlay of the defect information of the first type requires at least 3 of the batches
  • the identification codes of the products are screened to obtain, and the defect overlay of the second type of defect information needs to be screened to obtain the identification codes of at least one batch of products.
  • the first type of defect information is one of a foreign substance on the gate layer film and a foreign substance on the semiconductor layer film; There is one of small foreign matters and semiconductor material residues.
  • the batch of products is one of a thin film transistor substrate, a glass substrate, a color film substrate, or a liquid crystal cell.
  • the method for determining aggregation of defect overlays further includes the following steps:
  • Yet another object of the present application is to provide a system for judging defect overlays.
  • a system for judging the aggregation of defect overlays includes:
  • the first screening module is used to screen out the defect group in the defect overlay, which is obtained by overlaying the defects of a batch of products with the same defect, at least two of the defect groups Same defect;
  • a first judgment module used to judge whether there is a defect group with the number of the same defects greater than a first preset value in the defect stack
  • defect overlay is aggregated
  • the first screening module includes:
  • a first calculation unit configured to calculate the distance between any two defects in the defect stack, if the distance between any two defects in the defect stack is less than a second preset value, then the Any two defects belong to the same defect group.
  • the second preset value is 280-320 microns.
  • the second preset value is 300 ⁇ m.
  • system for judging defect overlapping image aggregation further includes:
  • the tracing module is used to trace the source of the defect overlay, and the tracing module includes:
  • the first identification unit is used to identify the identification code of the batch of products corresponding to the accumulation of defective overlay images to trace the processing equipment of the batch of products corresponding to the accumulation of defective overlay images, the identification code
  • the processing equipment includes a single-path machine and a reaction chamber.
  • the traceability module includes:
  • the first response unit is used for responding to extracting the batch produced by the single-path machine from the defective stack when the processing equipment of the batch of products corresponding to the stack of defective stacks is a single-path machine Overlap the defects of the secondary products, and simultaneously issue the alarm message that the single-path machine causes the accumulation of defective overlays;
  • the second response unit is used for responding to extracting the batch of products produced by the reaction chamber from the defect overlay when the processing equipment of the batch of products corresponding to the aggregation of the defect overlay is the reaction chamber.
  • the defects are superimposed, and an alarm message is generated that the reaction chamber causes the defects to accumulate.
  • the defect information includes defect information of the first type and defect information of the second type, and the defect overlay of the defect information of the first type requires at least 3 of the batches
  • the identification codes of the products are screened to obtain, and the defect overlay of the second type of defect information needs to be screened to obtain the identification codes of at least one batch of products.
  • the first type of defect information is one of foreign matter on the gate layer film and foreign matter on the semiconductor layer film;
  • the second type of defect signal is before the coating process There is one of small foreign matters and semiconductor material residues.
  • the batch of products is one of a thin film transistor substrate, a glass substrate, a color film substrate, or a liquid crystal cell.
  • system for judging defect overlapping image aggregation further includes:
  • Output module used to output batch products
  • a second screening module for screening the output identification codes of the batch of products, the identification codes recording defect information of the batch of products;
  • the overlay module is used to overlay the defects of the batch of products with the same identification code to obtain the defect overlay.
  • This application proposes a method and system for judging the aggregation of defect overlays, and selecting the defect group in the defect overlay by an automated method to determine whether there is a defect group with a defect number greater than the first preset value in the defect overlay, thereby To determine whether there is aggregation of defect overlays, compared to manually observing whether the defect overlays are clustered, the present application found that the phenomenon of defect overlays aggregation is faster and more accurate.
  • FIG. 1 is a method for judging a defect overlay in the first embodiment of the present application
  • FIG. 2 is a method for judging the stacking of defects according to the second embodiment of the present application.
  • FIG. 3 is a system for judging defect overlays in the first embodiment of the present application.
  • FIG. 1 it is a method for judging a defect overlay in the first embodiment of the present application, and includes the following steps:
  • S10 Screen out the defect group in the defect overlay, the defect overlay is obtained by overlaying the defects of a batch of products with the same defect, and there are at least two of the same defects in the defect group;
  • the batch product is a thin film transistor substrate with defects.
  • the batch product may be a glass substrate with defects, a color filter substrate with defects, or a liquid crystal cell with defects.
  • the defect image includes the position information of the defect, and then process the defect image through the image processing device, such as a computer, etc., and overlay the defect images of the product with the same defect batch Defect overlay, which can reflect the location information of the same defect on the batch of products.
  • the defect overlay When the number of defects in the defect overlay is sufficient, multiple defect groups will be formed in the defect overlay, and the defect group containing multiple defects The probability that the same defect gathers in a certain area is greater, and the defect group with fewer defects will reflect the same defect in a certain area.
  • the size of the first preset value can be adjusted according to the number of batches of products, which is not specifically limited in this application.
  • the above solution selects the defect group in the defect overlay by an automated method, and determines whether there are defect groups with a defect number greater than the first preset value in the defect overlay, so as to automatically determine whether the defect overlay is clustered, as compared with manual observation Whether the overlays are clustered, this application found that the defect overlays clustering is faster and more accurate.
  • FIG. 2 it is a method for judging a defect overlay in the second embodiment of the present application, and includes the following steps:
  • the batch of products is a thin-film transistor substrate with defects, and the thin-film transistor substrate with defects is output in units of batches by the transfer device.
  • the defects on the thin film transistor substrate come from various processes for preparing the thin film transistor substrate, the types of defects are different, and the probability of different defects appearing on the thin film transistor is also different.
  • a batch of products is output to prepare for the subsequent defect overlay.
  • each defective product When a batch of products is output, each defective product has an identification code, which records the defect information of the defective product, and the identification code is screened by using a scanning device to screen out products with the same defect.
  • the identification code For the first type of defects with a low probability of occurrence, such as foreign objects on the gate layer film and foreign objects on the semiconductor layer film, at least three batches of product identification codes need to be identified to obtain the defect overlay.
  • the occurrence probability Larger second-type defects, such as small foreign objects and semiconductor material residues before the coating process, need to identify at least one batch of product identification codes to obtain a defect overlay.
  • the image of the defect of the product with the same defect is acquired by the imaging device, and the image includes the position information of the defect, and then the defective image is processed by the image processing device, such as a computer, etc., will have the same
  • the defect image of the defective batch product is overlaid to obtain the defect overlay.
  • the defect overlay can reflect the position distribution information of the same defect on the batch of products. When the number of defects in the defect overlay is sufficient, it will be formed in the defect overlay. For multiple defect groups, a defect group containing multiple defects has a higher probability of reflecting the same defects in a certain area, and a defect group containing fewer defects has a lower probability of reflecting the same defects in a certain area.
  • the same defects on different thin film transistor substrates are located at different positions. By overlaying the same defects in different positions to obtain a defect overlay, and then calculating the distance of any two defects in the defect overlay to define any two Whether the defects belong to the same defect group. Since the size of general defects is in the micron level, that is, the size of the defects is small, it can be considered that such defects are point defects.
  • the second preset value is 280-320 microns. Specifically, the size of the second preset value can be set to 300 Micrometer, according to the second preset value to determine whether the two defects belong to the same defect group, in other embodiments, the second preset value can be adjusted according to the actual situation, for larger line defects or surface defects can be directly It is judged as aggregation.
  • defect overlay If there are defect groups with the same number of defects greater than the first preset value in the defect overlay, the defect overlay is determined to be aggregated;
  • defect overlay If there is no defect group with the same number of defects greater than the first preset value in the defect overlay, it is determined that the defect overlay is not aggregated.
  • the identification code Identify the identification code of the batch of products corresponding to the stacking of defects, to trace the processing equipment of the batch of products corresponding to the stacking of defects, the identification code records the defect information of the batch of products, the processing equipment includes a single path Machine and reaction chamber;
  • the processing equipment includes a single-path machine and a reaction chamber.
  • the single-path machine includes a physical deposition device and a wet etching device.
  • the reaction chamber includes a chemical reaction chamber and a reaction chamber for dry etching, that is, the identification code of the defective product can be adjusted. Taking the production and processing information of the defective product, combined with the defect information reflected in the identification code, you can trace the processing equipment that caused the defect.
  • the processing equipment of the batch of products corresponding to the stacking of defective stacks is a single-path machine
  • the defects of the batch of products produced by the single-path machine are extracted from the defective stack and stacked
  • the alarm message that the single-path machine caused the accumulation of defect overlays is issued at the same time;
  • the processing equipment of the batch of products corresponding to the stacking of defective stacks is a reaction chamber
  • the defects of the batch of products produced in the reaction chamber are extracted from the stack of defectives and stacked, and the Describe the alarm message in the reaction chamber that causes defects to stack up.
  • the processing equipment After tracing to the processing equipment that caused the defect, extract the defects caused by the same processing equipment from the defect stack and use the overlay to serve as the history information of the defects caused by the processing equipment, and at the same time send out the alarm information that the processing equipment caused the stack of defects.
  • the processing equipment can avoid further product defects.
  • the above embodiment determines whether the defect overlays are aggregated through an automated method, and then traces the cause of the defect overlays, which is helpful for solving the causes of the defect overlays.
  • the present application also provides a system 30 for judging defect overlays, including:
  • the first screening module 31 is used to screen out defect groups in a defect overlay, which is obtained by overlaying defects of a batch of products with the same defect, and there are at least two identical defects in the defect group;
  • the first judgment module 32 is used to judge whether there is a defect group with the number of the same defect greater than the first preset value in the defect overlay,
  • the defect overlay is determined to be aggregated
  • the first screening module includes:
  • the first calculation unit is used to calculate the distance between any two defects in the defect stack, if the distance between any two defects in the defect stack is less than the second preset value, then any two defects belong to the same A defect group.
  • system for determining the aggregation of defect overlays also includes:
  • the traceability module is used for tracing the source of defect overlays.
  • the traceability module includes:
  • the first identification unit is used to identify the identification code of the batch of products corresponding to the stacking of defects, to trace the processing equipment of the batch of products corresponding to the stacking of defects, and the identification code records the defects of the batch of products Information
  • processing equipment includes single-path machines and reaction chambers.
  • the traceability module includes:
  • the first response unit is used to respond to if the processing equipment of the batch of products corresponding to the accumulation of the defect stack is a single-path machine, then extract the defects of the batch of products produced by the single-path machine from the defect stack and perform Overlay, and simultaneously send out the warning message that the single-path machine causes the defective overlay to gather;
  • the second response unit is used to respond to if the processing equipment of the batch of products corresponding to the accumulation of the defect overlay is the reaction chamber, then extract the defects of the batch of products produced in the reaction chamber from the defect overlay and perform the overlay. And it sends out the warning message that the reaction room causes the defects to gather together.
  • the defect information includes the first type of defect information and the second type of defect information.
  • the defect overlay of the first type of defect information requires identification of at least three batches of product identification codes to obtain the defect of the second type of defect information Overlays need to identify at least one batch of product identification codes to obtain.
  • system for determining the aggregation of defect overlays also includes:
  • Output module used to output batch products
  • a second screening module for screening the output identification codes of the batch of products, the identification codes recording defect information of the batch of products;
  • the overlay module is used to overlay the defects of the batch of products with the same identification code to obtain the defect overlay.
  • the system for judging the defect overlay in this application selects the defect group in the defect overlay by an automated method, and determines whether there is a defect group with a defect number greater than the first preset value in the defect overlay, thereby determining whether the defect overlay Clustering, as compared to manually observing whether the defect overlay is clustered, this application found that the defect overlay is faster and more accurate.
  • the traceability module traces the processing equipment caused by the defect overlay and issues related alarm information, which is beneficial to solve the causes of the defect overlay.
  • the system for judging defect overlays provided in the above embodiments is an example based on the division of the above functional modules.
  • the above functions can be allocated by different functional modules according to needs, that is, to determine defects
  • the system of stacking graphs can be divided into different functional modules to complete all or part of the functions described above.

Landscapes

  • Engineering & Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Artificial Intelligence (AREA)
  • Evolutionary Biology (AREA)
  • Evolutionary Computation (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Testing Or Measuring Of Semiconductors Or The Like (AREA)
  • Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)

Abstract

一种判断缺陷叠图聚集的方法及其系统（30），通过自动化的方法筛选出缺陷叠图中的缺陷群，判断缺陷叠图中是否存在缺陷数目大于第一预设值的缺陷群，从而判断缺陷叠图是否存在聚集，相对于人工观察缺陷叠图是否聚集，缺陷叠图聚集现象的速度更快且更准确。

Description

判断缺陷叠图聚集的方法及其系统 技术领域
本申请涉及显示器制造技术领域，尤其涉及一种判断缺陷叠图聚集的方法及其系统。
 背景技术
显示器生产过程中，一些缺陷会产生会集中在产品的固定区域内，因此通过对缺陷进行叠图会发现缺陷聚集的趋势，当发现某种缺陷具有聚集趋势时，此类异常需要及时解决，不然将导致后续生产的产品继续出现同样的异常从而导致更大范围的产品缺陷。目前，主要通过人员手动对产品上存在的缺陷进行叠图并观察缺陷叠图的结果来判断是否存在缺陷聚集现象，然而由于人员的失误会导致异常发现不及时问题。
因此，有必要提供一种技术方案以解决现有技术由于不能自动化判断缺陷叠图是否存在缺陷聚集现象导致异常发现不及时的问题。
 技术问题
本申请的目的在于提供一种判断缺陷叠图聚集的方法，以实现自动化地判断缺陷叠图聚集，从而快速地发现缺陷叠图聚集现象。
 技术解决方案
一种判断缺陷叠图聚集的方法，所述判断缺陷叠图聚集的方法包括如下步骤：
筛选出缺陷叠图中的缺陷群，所述缺陷叠图是将具有相同缺陷的批次产品的缺陷进行叠图得到的，所述缺陷群中有至少两个所述相同缺陷；
若所述缺陷叠图中存在所述相同缺陷的数目大于第一预设值的缺陷群，则判断所述缺陷叠图为聚集；
若所述缺陷叠图中不存在所述相同缺陷的数目大于所述第一预设值的缺陷群，则判断所述缺陷叠图为未聚集。
在上述判断缺陷叠图聚集的方法中，所述筛选出所述缺陷叠图中的缺陷群包括如下步骤：
若所述缺陷叠图中的任意两个缺陷之间的距离小于第二预设值，则所述任意两个缺陷属于同一个缺陷群。
在上述判断缺陷叠图聚集的方法中，所述第二预设值为280-320微米。
在上述判断缺陷叠图聚集的方法中，所述第二预设值为300微米。
在上述判断缺陷叠图聚集的方法中，所述判断缺陷叠图聚集的方法还包括追溯缺陷叠图聚集的来源，所述追溯缺陷叠图聚集的来源包括如下步骤：
对所述判断为缺陷叠图聚集所对应的批次产品的识别码进行识别，以追溯所述缺陷叠图聚集所对应的批次产品的加工设备，所述识别码记录所述批次产品的缺陷信息，所述加工设备包括单路径机台及反应腔。
在上述判断缺陷叠图聚集的方法中，若所述缺陷叠图聚集所对应的批次产品的加工设备为单路径机台，则从所述缺陷叠图中提取出所述单路径机台所生产的批次产品的缺陷并进行叠图，同时发出所述单路径机台造成缺陷叠图聚集的警报信息；
若所述缺陷叠图聚集所对应的批次产品的加工设备为反应室，则从所述缺陷叠图中提取出所述反应室所生产的批次产品的缺陷并进行叠图，并发出所述反应室造成缺陷叠图聚集的警报信息。
在上述判断缺陷叠图聚集的方法中，所述缺陷信息包括第一类缺陷信息和第二类缺陷信息，所述第一类缺陷信息的所述缺陷叠图需要对至少3个所述批次产品的所述识别码进行筛选以得到，所述第二类缺陷信息的所述缺陷叠图需要对至少一个批次产品的所述识别码进行筛选以得到。
在上述判断缺陷叠图聚集的方法中，所述第一类缺陷信息为栅极层膜上有异物、半导体层膜上有异物中的一种；所述第二类缺陷信号为涂布工艺前有小异物、半导体材料残留中的一种。
在上述判断缺陷叠图聚集的方法中，所述批次产品为薄膜晶体管基板、玻璃基板、彩膜基板或液晶盒中的一种。
在上述判断缺陷叠图聚集的方法中，所述判断缺陷叠图聚集的方法还包括如下步骤：
输出批次产品；
对输出的所述批次产品的识别码进行筛选，所述识别码记录所述批次产品的缺陷信息；
对具有相同所述识别码的所述批次产品的缺陷进行叠图，得所述缺陷叠图。
本申请的又一目的是提供一种判断缺陷叠图聚集的系统。
为实现上述目的，技术方案如下。
一种判断缺陷叠图聚集的系统，所述判断缺陷叠图聚集的系统包括：
第一筛选模块，用于筛选出缺陷叠图中的缺陷群，所述缺陷叠图是将具有相同缺陷的批次产品的缺陷进行叠图得到的，所述缺陷群中有至少两个所述相同缺陷；
第一判断模块，用于判断所述缺陷叠图中是否存在所述相同缺陷的数目大于第一预设值的缺陷群，
若所述缺陷叠图中存在所述相同缺陷的数目大于第一预设值的缺陷群，则判断所述缺陷叠图为聚集；
若所述缺陷叠图中不存在所述相同缺陷的数目大于所述第一预设值的缺陷群，则判断所述缺陷叠图为未聚集。
在上述判断缺陷叠图聚集的系统中，所述第一筛选模块包括：
第一计算单元，用于计算所述缺陷叠图中的任意两个缺陷之间的距离，若所述缺陷叠图中的任意两个缺陷之间的距离小于第二预设值，则所述任意两个缺陷属于同一个缺陷群。
在上述判断缺陷叠图聚集的系统中，所述第二预设值为280-320微米。
在上述判断缺陷叠图聚集的系统中，所述第二预设值为300微米。
在上述判断缺陷叠图聚集的系统中，所述判断缺陷叠图聚集的系统还包括：
追溯模块，用于追溯缺陷叠图聚集的来源，所述追溯模块包括：
第一识别单元，用于对所述判断为缺陷叠图聚集所对应的批次产品的识别码进行识别，以追溯所述缺陷叠图聚集所对应的批次产品的加工设备，所述识别码记录所述批次产品的缺陷信息，所述加工设备包括单路径机台及反应腔。
在上述判断缺陷叠图聚集的系统中，所述追溯模块包括:
第一响应单元，用于响应若所述缺陷叠图聚集所对应的批次产品的加工设备为单路径机台时，则从所述缺陷叠图中提取出所述单路径机台所生产的批次产品的缺陷并进行叠图，同时发出所述单路径机台造成缺陷叠图聚集的警报信息；
第二响应单元，用于响应若所述缺陷叠图聚集所对应的批次产品的加工设备为反应室时，则从所述缺陷叠图中提取出所述反应室所生产的批次产品的缺陷并进行叠图，并发出所述反应室造成缺陷叠图聚集的警报信息。
在上述判断缺陷叠图聚集的系统中，所述缺陷信息包括第一类缺陷信息和第二类缺陷信息，所述第一类缺陷信息的所述缺陷叠图需要对至少3个所述批次产品的所述识别码进行筛选以得到，所述第二类缺陷信息的所述缺陷叠图需要对至少一个批次产品的所述识别码进行筛选以得到。
在上述判断缺陷叠图聚集的系统中，所述第一类缺陷信息为栅极层膜上有异物、半导体层膜上有异物中的一种；所述第二类缺陷信号为涂布工艺前有小异物、半导体材料残留中的一种。
在上述判断缺陷叠图聚集的系统中，所述批次产品为薄膜晶体管基板、玻璃基板、彩膜基板或液晶盒中的一种。
在上述判断缺陷叠图聚集的系统中，所述判断缺陷叠图聚集的系统还包括：
输出模块，用于输出批次产品；
第二筛选模块，用于对输出的所述批次产品的识别码进行筛选，所述识别码记录所述批次产品的缺陷信息；
叠图模块，用于对具有相同所述识别码的所述批次产品的缺陷进行叠图，得所述缺陷叠图。
 有益效果
本申请提出一种判断缺陷叠图聚集的方法及其系统，通过自动化的方法筛选出缺陷叠图中的缺陷群，判断缺陷叠图中是否存在缺陷数目大于第一预设值的缺陷群，从而判断缺陷叠图是否存在聚集，相对于人工观察缺陷叠图是否聚集，本申请发现缺陷叠图聚集现象的速度更快且更准确。
 附图说明
图1为本申请第一实施例的判断缺陷叠图聚集的方法；
图2为本申请第二实施例的判断缺陷叠图聚集的方法；
图3为本申请第一实施例的判断缺陷叠图聚集的系统。
 本发明的实施方式
以下各实施例的说明是参考附加的图示，用以例示本申请可用以实施的特定实施例。本申请所提到的方向用语，例如[上]、[下]、[前]、[后]、[左]、[右]、[内]、[外]、[侧面]等，仅是参考附加图式的方向。因此，使用的方向用语是用以说明及理解本申请，而非用以限制本申请。在图中，结构相似的单元是用以相同标号表示。
如图1所示，其为本申请第一实施例的判断缺陷叠图聚集的方法，包括如下步骤：
S10：筛选出缺陷叠图中的缺陷群，缺陷叠图是将具有相同缺陷的批次产品的缺陷进行叠图得到的，缺陷群中有至少两个所述相同缺陷；
在本实施例中，批次产品为具有缺陷的薄膜晶体管基板，在其他实施例中，批次产品可以为具有缺陷的玻璃基板、具有缺陷的彩膜基板或者具有缺陷的液晶盒等。通过成像设备获取批次产品上的缺陷图像，该缺陷图像包括缺陷的位置信息，再使缺陷图像通过图像处理设备处理，例如计算机等，将具有相同缺陷批次产品的缺陷图像进行叠图以得到缺陷叠图，缺陷叠图能够反应相同缺陷在批次产品上的位置分布信息，缺陷叠图中的缺陷数量足够多时，在缺陷叠图中会形成多个缺陷群，含有多个缺陷的缺陷群反应相同缺陷聚集在一定区域的概率较大，含有较少缺陷的缺陷群反应相同缺陷聚集在一定区域的概率较小。
S11：若缺陷叠图中存在所述相同缺陷的数目大于第一预设值的缺陷群，则判断缺陷叠图为聚集；
若缺陷叠图中不存在相同缺陷的数目大于第一预设值的缺陷群，则判断缺陷叠图为未聚集。
通过比较含有最多缺陷数目的缺陷群的缺陷数与第一预设值的大小，以判断缺陷是否在一定的区域聚集，从而判断缺陷叠图是否存在聚集的现象，如果存在缺陷数目大于第一预设值的缺陷群，则缺陷叠图中存在缺陷在一定区域聚集的现象。第一预设值的大小可以根据批次产品的数目进行调节，本申请不做具体的限定。
上述方案通过自动化的方法筛选出缺陷叠图中的缺陷群，判断缺陷叠图中是否存在缺陷数目大于第一预设值的缺陷群，从而自动化判断缺陷叠图是否存在聚集，相对于人工观察缺陷叠图是否聚集，本申请发现缺陷叠图聚集现象的速度更快且更准确。
如图2所示，其为本申请第二实施例的判断缺陷叠图聚集的方法，包括如下步骤：
S20：输出批次产品。
如第一实施例中所述，批次产品为具有缺陷的薄膜晶体管基板，通过传输装置以批次为单位输出具有缺陷的薄膜晶体管基板。薄膜晶体管基板上的缺陷来源于制备薄膜晶体管基板的各个制程，缺陷的种类各异，且不同的缺陷在薄膜晶体管上出现的概率也不同。本实施例通过输出批次产品以为后续得到缺陷叠图作准备。
S21：对输出的批次产品的识别码进行筛选。
批次产品输出时，每个缺陷产品上都具有识别码，该识别码记录缺陷产品的缺陷信息，通过使用扫描设备对识别码进行筛选，将具有相同缺陷的产品筛选出来。对于出现概率较小的第一类缺陷，例如栅极层膜上有异物、半导体层膜上有异物等，需要对至少3个批次产品的识别码进行识别才能得到缺陷叠图，对于出现概率较大的第二类缺陷，例如涂布工艺前有小异物、半导体材料残留等，需要对至少一个批次产品的识别码进行识别才能得到缺陷叠图。
S22：对具有相同识别码的批次产品的缺陷进行叠图，得缺陷叠图。
具体地，具有相同缺陷的产品筛选出来后，通过成像设备获取相同缺陷的产品的缺陷的图像，该图像包括缺陷的位置信息，再使缺陷图像通过图像处理设备处理，例如计算机等，将具有相同缺陷批次产品的缺陷图像进行叠图以得到缺陷叠图，缺陷叠图能够反应相同缺陷在批次产品上的位置分布信息，缺陷叠图中的缺陷数量足够多时，在缺陷叠图中会形成多个缺陷群，含有多个缺陷的缺陷群反应相同缺陷聚集在一定区域的概率较大，含有较少缺陷的缺陷群反应相同缺陷聚集在一定区域的概率较小。
S23：筛选出缺陷叠图中的缺陷群，若缺陷叠图中的任意两个缺陷之间的距离小于第二预设值，则任意两个缺陷属于同一个缺陷群。
不同薄膜晶体管基板上的相同缺陷位于不同的位置，通过将处于不同位置的相同缺陷进行叠图以得到缺陷叠图，再计算缺陷叠图中的任意两个缺陷的距离，以定义任意两个的缺陷是否属于同一缺陷群。由于一般缺陷的尺寸在微米级，即缺陷的尺寸较小，可以认为此类缺陷为点缺陷，第二预设值为280-320微米，具体地，第二预设值的大小可以设置为300微米，根据第二预设值以判断两个缺陷是否属于同一个缺陷群，在其他实施例中，第二预设值可以根据实际情况进行调整，对于尺寸较大的线缺陷或者面缺陷可以直接判断为为聚集。
S24：若缺陷叠图中存在相同缺陷的数目大于第一预设值的缺陷群，则判断缺陷叠图为聚集；
若缺陷叠图中不存在相同缺陷的数目大于所述第一预设值的缺陷群中，则判断缺陷叠图为未聚集。
S25：追溯缺陷叠图聚集的来源，包括：
对判断为缺陷叠图聚集所对应的批次产品的识别码进行识别，以追溯缺陷叠图聚集所对应的批次产品的加工设备，识别码记录批次产品的缺陷信息，加工设备包括单路径机台及反应腔；
对判定为缺陷聚集的批次产品的识别码进行识别，然后根据识别码可以追溯识别码对应的产品的生产加工信息，另外，根据识别码所反应的缺陷信息可以追溯到造成缺陷的加工设备。加工设备包括单路径机台及反应腔，单路径机台又包括物理沉积装置及湿蚀刻装置等，反应腔包括化学反应腔以及干蚀刻用的反应腔，即，通过缺陷产品的识别码可以调取缺陷产品的生产加工信息，结合识别码反应出的缺陷信息，可以追踪造成缺陷的加工设备。
接着，若所述缺陷叠图聚集所对应的批次产品的加工设备为单路径机台，则从所述缺陷叠图中提取出所述单路径机台所生产的批次产品的缺陷并进行叠图，同时发出所述单路径机台造成缺陷叠图聚集的警报信息；
若所述缺陷叠图聚集所对应的批次产品的加工设备为反应室，则从所述缺陷叠图中提取出所述反应室所生产的批次产品的缺陷并进行叠图，并发出所述反应室造成缺陷叠图聚集的警报信息。
追踪到造成缺陷的加工设备后，从缺陷叠图中提取出同一加工设备造成的缺陷并进行叠图以作为该加工设备造成的缺陷历史信息，同时发出该加工设备造成缺陷叠图聚集的警报信息，通过对造成缺陷叠图的加工设备进行检修，可以避免加工设备进一步造成产品缺陷。
上述实施例通过自动化的方法判断出缺陷叠图是否聚集，再对缺陷叠图聚集的原因进行追溯，有利于对造成缺陷叠图聚集的原因进行解决。
如图3所示，本申请还提供一种判断缺陷叠图聚集的系统30，包括：
第一筛选模块31，用于筛选出缺陷叠图中的缺陷群，缺陷叠图是将具有相同缺陷的批次产品的缺陷进行叠图得到的，缺陷群中有至少两个相同缺陷；
第一判断模块32，用于判断缺陷叠图中是否存在相同缺陷的数目大于第一预设值的缺陷群，
若缺陷叠图中存在相同缺陷的数目大于第一预设值的缺陷群，则判断缺陷叠图为聚集；
若缺陷叠图中不存在相同缺陷的数目大于第一预设值的缺陷群，则判断所述缺陷叠图为未聚集。
进一步地，第一筛选模块包括：
第一计算单元，用于计算缺陷叠图中的任意两个缺陷之间的距离，若缺陷叠图中的任意两个缺陷之间的距离小于第二预设值，则任意两个缺陷属于同一个缺陷群。
进一步地，判断缺陷叠图聚集的系统还包括：
追溯模块，用于追溯缺陷叠图聚集的来源，追溯模块包括：
第一识别单元，用于对判断为缺陷叠图聚集所对应的批次产品的识别码进行识别，以追溯缺陷叠图聚集所对应的批次产品的加工设备，识别码记录批次产品的缺陷信息，加工设备包括单路径机台及反应腔。
进一步地，追溯模块包括:
第一响应单元，用于响应若缺陷叠图聚集所对应的批次产品的加工设备为单路径机台时，则从缺陷叠图中提取出单路径机台所生产的批次产品的缺陷并进行叠图，同时发出单路径机台造成缺陷叠图聚集的警报信息；
第二响应单元，用于响应若缺陷叠图聚集所对应的批次产品的加工设备为反应室时，则从缺陷叠图中提取出反应室所生产的批次产品的缺陷并进行叠图，并发出反应室造成缺陷叠图聚集的警报信息。
进一步地，缺陷信息包括第一类缺陷信息和第二类缺陷信息，第一类缺陷信息的缺陷叠图需要对至少3个批次产品的识别码进行识别以得到，第二类缺陷信息的缺陷叠图需要对至少一个批次产品的识别码进行识别以得到。
进一步地，判断缺陷叠图聚集的系统还包括：
输出模块，用于输出批次产品；
第二筛选模块，用于对输出的所述批次产品的识别码进行筛选，所述识别码记录所述批次产品的缺陷信息；
叠图模块，用于对具有相同所述识别码的所述批次产品的缺陷进行叠图，得所述缺陷叠图。
本申请的判断缺陷叠图聚集的系统通过自动化的方法筛选出缺陷叠图中的缺陷群，判断缺陷叠图中是否存在缺陷数目大于第一预设值的缺陷群，从而判断缺陷叠图是否存在聚集，相对于人工观察缺陷叠图是否聚集，本申请发现缺陷叠图聚集现象的速度更快且更准确。此外，通过追溯模块对缺陷叠图造成的加工设备进行追溯并发出相关警报信息，有利于对造成缺陷叠图聚集的原因进行解决。
需要说明的是：上述实施例提供的判断缺陷叠图聚集的系统是根据上述各功能模块的划分进行举例，实际应用中，可以根据需要而将上述功能分配由不同的功能模块完成，即判断缺陷叠图聚集的系统可以划分成不同的功能模块，以完成上述描述的全部或部分功能。
综上所述，虽然本申请已以优选实施例揭露如上，但上述优选实施例并非用以限制本申请，本领域的普通技术人员，在不脱离本申请的精神和范围内，均可作各种更动与润饰，因此本申请的保护范围以权利要求界定的范围为准。

Claims (20)

  1. 一种判断缺陷叠图聚集的方法，其中，所述判断缺陷叠图聚集的方法包括如下步骤：
    筛选出缺陷叠图中的缺陷群，所述缺陷叠图是将具有相同缺陷的批次产品的缺陷进行叠图得到的，所述缺陷群中有至少两个所述相同缺陷；
    若所述缺陷叠图中存在所述相同缺陷的数目大于第一预设值的缺陷群，则判断所述缺陷叠图为聚集；
    若所述缺陷叠图中不存在所述相同缺陷的数目大于所述第一预设值的缺陷群，则判断所述缺陷叠图为未聚集。
  2. 根据权利要求1所述的判断缺陷叠图聚集的方法，其中，所述筛选出所述缺陷叠图中的缺陷群包括如下步骤：
    若所述缺陷叠图中的任意两个缺陷之间的距离小于第二预设值，则所述任意两个缺陷属于同一个缺陷群。
  3. 根据权利要求2所述的判断缺陷叠图聚集的方法，其中，所述第二预设值为280-320微米。
  4. 根据权利要求3所述的判断缺陷叠图聚集的方法，其中，所述第二预设值为300微米。
  5. 根据权利要求1所述的判断缺陷叠图聚集的方法，其中，所述判断缺陷叠图聚集的方法还包括追溯缺陷叠图聚集的来源，所述追溯缺陷叠图聚集的来源包括如下步骤：
    对所述判断为缺陷叠图聚集所对应的批次产品的识别码进行识别，以追溯所述缺陷叠图聚集所对应的批次产品的加工设备，所述识别码记录所述批次产品的缺陷信息，所述加工设备包括单路径机台及反应腔。
  6. 根据权利要求5所述的判断缺陷叠图聚集的方法，其中，若所述缺陷叠图聚集所对应的批次产品的加工设备为单路径机台，则从所述缺陷叠图中提取出所述单路径机台所生产的批次产品的缺陷并进行叠图，同时发出所述单路径机台造成缺陷叠图聚集的警报信息；
    若所述缺陷叠图聚集所对应的批次产品的加工设备为反应室，则从所述缺陷叠图中提取出所述反应室所生产的批次产品的缺陷并进行叠图，并发出所述反应室造成缺陷叠图聚集的警报信息。
  7. 根据权利要求5所述的判断缺陷叠图聚集的方法，其中，所述缺陷信息包括第一类缺陷信息和第二类缺陷信息，所述第一类缺陷信息的所述缺陷叠图需要对至少3个所述批次产品的所述识别码进行筛选以得到，所述第二类缺陷信息的所述缺陷叠图需要对至少一个批次产品的所述识别码进行筛选以得到。
  8. 根据权利要求7所述的判断缺陷叠图聚集的方法，其中，所述第一类缺陷信息为栅极层膜上有异物、半导体层膜上有异物中的一种；所述第二类缺陷信号为涂布工艺前有小异物、半导体材料残留中的一种。
  9. 根据权利要求1所述的判断缺陷叠图聚集的方法，其中，所述批次产品为薄膜晶体管基板、玻璃基板、彩膜基板或液晶盒中的一种。
  10. 根据权利要求1所述的判断缺陷叠图聚集的方法，其中，所述判断缺陷叠图聚集的方法还包括如下步骤：
    输出批次产品；
    对输出的所述批次产品的识别码进行筛选，所述识别码记录所述批次产品的缺陷信息；
    对具有相同所述识别码的所述批次产品的缺陷进行叠图，得所述缺陷叠图。
  11. 一种判断缺陷叠图聚集的系统，其中，所述判断缺陷叠图聚集的系统包括：
    第一筛选模块，用于筛选出缺陷叠图中的缺陷群，所述缺陷叠图是将具有相同缺陷的批次产品的缺陷进行叠图得到的，所述缺陷群中有至少两个所述相同缺陷；
    第一判断模块，用于判断所述缺陷叠图中是否存在所述相同缺陷的数目比第一预设值大的缺陷群，
    若所述缺陷叠图中存在所述相同缺陷的数目大于第一预设值的缺陷群，则判断所述缺陷叠图为聚集；
    若所述缺陷叠图中不存在所述相同缺陷的数目大于所述第一预设值的缺陷群，则判断所述缺陷叠图为未聚集。
  12. 根据权利要求11所述的判断缺陷叠图聚集的系统，其中，所述第一筛选模块包括：
    第一计算单元，用于计算所述缺陷叠图中的任意两个缺陷之间的距离，若所述缺陷叠图中的任意两个缺陷之间的距离小于第二预设值，则所述任意两个缺陷属于同一个缺陷群。
  13. 根据权利要求12所述的判断缺陷叠图聚集的系统，其中，所述第二预设值为280-320微米。
  14. 根据权利要求13所述的判断缺陷叠图聚集的系统，其中，所述第二预设值为300微米。
  15. 根据权利要求11所述的判断缺陷叠图聚集的系统，其中，所述判断缺陷叠图聚集的系统还包括：
    追溯模块，用于追溯缺陷叠图聚集的来源，所述追溯模块包括：
    第一识别单元，用于对所述判断为缺陷叠图聚集所对应的批次产品的识别码进行识别，以追溯所述缺陷叠图聚集所对应的批次产品的加工设备，所述识别码记录所述批次产品的缺陷信息，所述加工设备包括单路径机台及反应腔。
  16. 根据权利要求15所述的判断缺陷叠图聚集的系统，其中，所述追溯模块包括:
    第一响应单元，用于响应若所述缺陷叠图聚集所对应的批次产品的加工设备为单路径机台时，则从所述缺陷叠图中提取出所述单路径机台所生产的批次产品的缺陷并进行叠图，同时发出所述单路径机台造成缺陷叠图聚集的警报信息；
    第二响应单元，用于响应若所述缺陷叠图聚集所对应的批次产品的加工设备为反应室时，则从所述缺陷叠图中提取出所述反应室所生产的批次产品的缺陷并进行叠图，并发出所述反应室造成缺陷叠图聚集的警报信息。
  17. 根据权利要求15所述的判断缺陷叠图聚集的系统，其中，所述缺陷信息包括第一类缺陷信息和第二类缺陷信息，所述第一类缺陷信息的所述缺陷叠图需要对至少3个所述批次产品的所述识别码进行筛选以得到，所述第二类缺陷信息的所述缺陷叠图需要对至少一个批次产品的所述识别码进行筛选以得到。
  18. 根据权利要求17所述的判断缺陷叠图聚集的系统，其中，所述第一类缺陷信息为栅极层膜上有异物、半导体层膜上有异物中的一种；所述第二类缺陷信号为涂布工艺前有小异物、半导体材料残留中的一种。
  19. 根据权利要求11所述的判断缺陷叠图聚集的系统，其中，所述批次产品为薄膜晶体管基板、玻璃基板、彩膜基板或液晶盒中的一种。
  20. 根据权利要求11所述的判断缺陷叠图聚集的系统，其中，所述判断缺陷叠图聚集的系统还包括：
    输出模块，用于输出批次产品；
    第二筛选模块，用于对输出的所述批次产品的识别码进行筛选，所述识别码记录所述批次产品的缺陷信息；
    叠图模块，用于对具有相同所述识别码的所述批次产品的缺陷进行叠图，得所述缺陷叠图。
PCT/CN2018/120979 2018-10-18 2018-12-13 判断缺陷叠图聚集的方法及其系统 WO2020077784A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201811212402.2 2018-10-18
CN201811212402.2A CN109522931A (zh) 2018-10-18 2018-10-18 判断缺陷叠图聚集的方法及其系统

Publications (1)

Publication Number Publication Date
WO2020077784A1 true WO2020077784A1 (zh) 2020-04-23

Family

ID=65772022

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2018/120979 WO2020077784A1 (zh) 2018-10-18 2018-12-13 判断缺陷叠图聚集的方法及其系统

Country Status (2)

Country Link
CN (1) CN109522931A (zh)
WO (1) WO2020077784A1 (zh)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110111265B (zh) * 2019-04-04 2021-08-06 Tcl华星光电技术有限公司 叠加图形的校正方法及叠加图形的校正装置
CN111403309A (zh) * 2020-03-30 2020-07-10 上海华力集成电路制造有限公司 晶圆聚集状缺陷检测方法及其检测系统

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101484910A (zh) * 2006-07-06 2009-07-15 旭硝子株式会社 聚类系统及缺陷种类判定装置
CN103064206A (zh) * 2013-01-08 2013-04-24 深圳市华星光电技术有限公司 玻璃基板的缺陷检测方法
JP2013164371A (ja) * 2012-02-13 2013-08-22 Nippon Electric Glass Co Ltd 蛍光体含有ガラス部材の検査装置及び検査方法
CN105205803A (zh) * 2015-08-18 2015-12-30 武汉华星光电技术有限公司 显示面板缺陷检测方法
CN106067427A (zh) * 2016-05-25 2016-11-02 上海华力微电子有限公司 局部曝光异常缺陷自动检测方法
CN107941812A (zh) * 2017-12-20 2018-04-20 联想(北京)有限公司 信息处理方法及电子设备

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2001230289A (ja) * 2000-02-15 2001-08-24 Hitachi Ltd 欠陥解析方法および欠陥解析システム
CN107767024A (zh) * 2017-09-20 2018-03-06 彩虹集团（邵阳）特种玻璃有限公司 一种用于盖板玻璃生产线产品质量实时监测系统及方法

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101484910A (zh) * 2006-07-06 2009-07-15 旭硝子株式会社 聚类系统及缺陷种类判定装置
JP2013164371A (ja) * 2012-02-13 2013-08-22 Nippon Electric Glass Co Ltd 蛍光体含有ガラス部材の検査装置及び検査方法
CN103064206A (zh) * 2013-01-08 2013-04-24 深圳市华星光电技术有限公司 玻璃基板的缺陷检测方法
CN105205803A (zh) * 2015-08-18 2015-12-30 武汉华星光电技术有限公司 显示面板缺陷检测方法
CN106067427A (zh) * 2016-05-25 2016-11-02 上海华力微电子有限公司 局部曝光异常缺陷自动检测方法
CN107941812A (zh) * 2017-12-20 2018-04-20 联想(北京)有限公司 信息处理方法及电子设备

Also Published As

Publication number Publication date
CN109522931A (zh) 2019-03-26

Similar Documents

Publication Publication Date Title
JP6917943B2 (ja) 欠陥検出方法、欠陥検出のためのシステムおよび訓練方法
US8041100B2 (en) System for specifying equipment causing failure
JP2002100660A (ja) 欠陥検出方法と欠陥観察方法及び欠陥検出装置
WO2014034349A1 (ja) 欠陥分類装置、欠陥分類方法、制御プログラム、および記録媒体
TW201702776A (zh) 光學薄膜製程即時監測系統及其方法
JP2007194262A (ja) 欠陥判定システムおよび基板処理システム
WO2020077784A1 (zh) 判断缺陷叠图聚集的方法及其系统
TW201909010A (zh) 設計佈局為主的快速線上缺陷診斷、分類及取樣方法及系統
JP2010165876A (ja) 欠陥関連付け装置、基板検査システム、および欠陥関連付け方法
US8472696B2 (en) Observation condition determination support device and observation condition determination support method
CN104716066B (zh) 一种侦测图形底部光刻胶残留的缺陷检测方法
JP2015004674A (ja) 光学フィルムの欠陥判別方法
US20220139744A1 (en) Wafer defect analysis method and system, device and medium
JP3665215B2 (ja) 異常原因特定システムおよびその方法
JP4858106B2 (ja) カラーフィルタの欠陥検査方法
WO2017049863A1 (zh) 液晶滴注系统及控制方法
JP2011128030A (ja) 欠陥検査装置、及び欠陥情報管理方法
JP2004063708A (ja) 欠陥発生状態の監視方法およびその装置
US8421585B2 (en) Alarm apparatus and manufacturing method
JP5604196B2 (ja) マクロ検査方法および検査システム
WO2024000356A1 (zh) 数据处理方法及装置、数据显示方法及装置、设备和介质
CN113012137B (zh) 一种面板缺陷检查方法、系统、终端设备及存储介质
CN110109945B (zh) 应用于基板的aoi检测方法、装置、存储介质及aoi检测设备
JP3750220B2 (ja) 半導体装置の製造方法
CN103887213A (zh) 一种调整检测频率的缺陷检测方法

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 18937055

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 18937055

Country of ref document: EP

Kind code of ref document: A1